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 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
290 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
291 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
292 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
293 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
294 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
295 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
296 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
297 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
298 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
302 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
303 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
304 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
305 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
306 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
307 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
308 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
310 # Apple HTTP Live Streaming
311 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
312 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
313 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
314 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
315 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
316 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
317 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
320 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
321 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
322 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
323 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
324 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
325 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
326 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
327 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
328 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
329 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
330 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
333 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
334 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
335 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
338 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
339 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
340 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
341 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
342 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
343 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
344 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
345 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
346 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
347 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
348 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
349 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
350 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
351 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
352 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
353 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
354 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
355 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
356 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
357 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
358 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
359 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
362 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
363 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
365 # Dash webm audio with opus inside
366 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
367 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
368 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
371 '_rtmp': {'protocol': 'rtmp'},
373 _SUBTITLE_FORMATS = ('ttml', 'vtt')
378 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
382 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
383 'uploader': 'Philipp Hagemeister',
384 'uploader_id': 'phihag',
385 'upload_date': '20121002',
386 '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 .',
387 'categories': ['Science & Technology'],
388 'tags': ['youtube-dl'],
390 'dislike_count': int,
396 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
397 'note': 'Test generic use_cipher_signature video (#897)',
401 'upload_date': '20120506',
402 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
403 'alt_title': 'I Love It (feat. Charli XCX)',
404 'description': 'md5:782e8651347686cba06e58f71ab51773',
405 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
406 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
407 'iconic ep', 'iconic', 'love', 'it'],
408 'uploader': 'Icona Pop',
409 'uploader_id': 'IconaPop',
410 'creator': 'Icona Pop',
414 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
415 'note': 'Test VEVO video with age protection (#956)',
419 'upload_date': '20130703',
420 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
421 'alt_title': 'Tunnel Vision',
422 'description': 'md5:64249768eec3bc4276236606ea996373',
423 'uploader': 'justintimberlakeVEVO',
424 'uploader_id': 'justintimberlakeVEVO',
425 'creator': 'Justin Timberlake',
430 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
431 'note': 'Embed-only video (#1746)',
435 'upload_date': '20120608',
436 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
437 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
438 'uploader': 'SET India',
439 'uploader_id': 'setindia',
444 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
445 'note': 'Use the first video ID in the URL',
449 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
450 'uploader': 'Philipp Hagemeister',
451 'uploader_id': 'phihag',
452 'upload_date': '20121002',
453 '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 .',
454 'categories': ['Science & Technology'],
455 'tags': ['youtube-dl'],
457 'dislike_count': int,
460 'skip_download': True,
464 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
465 'note': '256k DASH audio (format 141) via DASH manifest',
469 'upload_date': '20121002',
470 'uploader_id': '8KVIDEO',
472 'uploader': '8KVIDEO',
473 'title': 'UHDTV TEST 8K VIDEO.mp4'
476 'youtube_include_dash_manifest': True,
480 # DASH manifest with encrypted signature
482 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
486 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
487 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
488 'uploader': 'AfrojackVEVO',
489 'uploader_id': 'AfrojackVEVO',
490 'upload_date': '20131011',
493 'youtube_include_dash_manifest': True,
497 # JS player signature function name containing $
499 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
503 'title': 'Taylor Swift - Shake It Off',
504 'alt_title': 'Shake It Off',
505 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
506 'uploader': 'TaylorSwiftVEVO',
507 'uploader_id': 'TaylorSwiftVEVO',
508 'upload_date': '20140818',
509 'creator': 'Taylor Swift',
512 'youtube_include_dash_manifest': True,
518 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
522 'upload_date': '20100909',
523 'uploader': 'The Amazing Atheist',
524 'uploader_id': 'TheAmazingAtheist',
525 'title': 'Burning Everyone\'s Koran',
526 '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',
529 # Normal age-gate video (No vevo, embed allowed)
531 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
535 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
536 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
537 'uploader': 'The Witcher',
538 'uploader_id': 'WitcherGame',
539 'upload_date': '20140605',
543 # Age-gate video with encrypted signature
545 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
549 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
550 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
551 'uploader': 'LloydVEVO',
552 'uploader_id': 'LloydVEVO',
553 'upload_date': '20110629',
557 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
559 'url': '__2ABJjxzNo',
563 'upload_date': '20100430',
564 'uploader_id': 'deadmau5',
565 'creator': 'deadmau5',
566 'description': 'md5:12c56784b8032162bb936a5f76d55360',
567 'uploader': 'deadmau5',
568 'title': 'Deadmau5 - Some Chords (HD)',
569 'alt_title': 'Some Chords',
571 'expected_warnings': [
572 'DASH manifest missing',
575 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
577 'url': 'lqQg6PlCWgI',
581 'upload_date': '20150827',
582 'uploader_id': 'olympic',
583 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
584 'uploader': 'Olympics',
585 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
588 'skip_download': 'requires avconv',
593 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
597 'stretched_ratio': 16 / 9.,
598 'upload_date': '20110310',
599 'uploader_id': 'AllenMeow',
600 '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',
602 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
605 # url_encoded_fmt_stream_map is empty string
607 'url': 'qEJwOuvDf7I',
611 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
613 'upload_date': '20150404',
614 'uploader_id': 'spbelect',
615 'uploader': 'Наблюдатели Петербурга',
618 'skip_download': 'requires avconv',
620 'skip': 'This live event has ended.',
622 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
624 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
628 'title': 'md5:7b81415841e02ecd4313668cde88737a',
629 'description': 'md5:116377fd2963b81ec4ce64b542173306',
630 'upload_date': '20150625',
631 'uploader_id': 'dorappi2000',
632 'uploader': 'dorappi2000',
633 'formats': 'mincount:33',
636 # DASH manifest with segment_list
638 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
639 'md5': '8ce563a1d667b599d21064e982ab9e31',
643 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
644 'uploader': 'Airtek',
645 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
646 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
647 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
650 'youtube_include_dash_manifest': True,
651 'format': '135', # bestvideo
655 # Multifeed videos (multiple cameras), URL is for Main Camera
656 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
659 'title': 'teamPGP: Rocket League Noob Stream',
660 'description': 'md5:dc7872fb300e143831327f1bae3af010',
666 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
667 'description': 'md5:dc7872fb300e143831327f1bae3af010',
668 'upload_date': '20150721',
669 'uploader': 'Beer Games Beer',
670 'uploader_id': 'beergamesbeer',
676 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
677 'description': 'md5:dc7872fb300e143831327f1bae3af010',
678 'upload_date': '20150721',
679 'uploader': 'Beer Games Beer',
680 'uploader_id': 'beergamesbeer',
686 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
687 'description': 'md5:dc7872fb300e143831327f1bae3af010',
688 'upload_date': '20150721',
689 'uploader': 'Beer Games Beer',
690 'uploader_id': 'beergamesbeer',
696 'title': 'teamPGP: Rocket League Noob Stream (zim)',
697 'description': 'md5:dc7872fb300e143831327f1bae3af010',
698 'upload_date': '20150721',
699 'uploader': 'Beer Games Beer',
700 'uploader_id': 'beergamesbeer',
704 'skip_download': True,
708 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
709 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
712 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
717 'url': 'http://vid.plus/FlRa-iH7PGw',
718 'only_matching': True,
721 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
722 # Also tests cut-off URL expansion in video description (see
723 # https://github.com/rg3/youtube-dl/issues/1892,
724 # https://github.com/rg3/youtube-dl/issues/8164)
725 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
729 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
730 'alt_title': 'Dark Walk',
731 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
732 'upload_date': '20151119',
733 'uploader_id': 'IronSoulElf',
734 'uploader': 'IronSoulElf',
735 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
738 'skip_download': True,
742 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
743 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
744 'only_matching': True,
747 # Video with yt:stretch=17:0
748 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
752 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
753 'description': 'md5:ee18a25c350637c8faff806845bddee9',
754 'upload_date': '20151107',
755 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
756 'uploader': 'CH GAMER DROID',
759 'skip_download': True,
763 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
764 'only_matching': True,
768 def __init__(self, *args, **kwargs):
769 super(YoutubeIE, self).__init__(*args, **kwargs)
770 self._player_cache = {}
772 def report_video_info_webpage_download(self, video_id):
773 """Report attempt to download video info webpage."""
774 self.to_screen('%s: Downloading video info webpage' % video_id)
776 def report_information_extraction(self, video_id):
777 """Report attempt to extract video information."""
778 self.to_screen('%s: Extracting video information' % video_id)
780 def report_unavailable_format(self, video_id, format):
781 """Report extracted video URL."""
782 self.to_screen('%s: Format %s not available' % (video_id, format))
784 def report_rtmp_download(self):
785 """Indicate the download will use the RTMP protocol."""
786 self.to_screen('RTMP download detected')
788 def _signature_cache_id(self, example_sig):
789 """ Return a string representation of a signature """
790 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
792 def _extract_signature_function(self, video_id, player_url, example_sig):
794 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
797 raise ExtractorError('Cannot identify player %r' % player_url)
798 player_type = id_m.group('ext')
799 player_id = id_m.group('id')
801 # Read from filesystem cache
802 func_id = '%s_%s_%s' % (
803 player_type, player_id, self._signature_cache_id(example_sig))
804 assert os.path.basename(func_id) == func_id
806 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
807 if cache_spec is not None:
808 return lambda s: ''.join(s[i] for i in cache_spec)
811 'Downloading player %s' % player_url
812 if self._downloader.params.get('verbose') else
813 'Downloading %s player %s' % (player_type, player_id)
815 if player_type == 'js':
816 code = self._download_webpage(
817 player_url, video_id,
819 errnote='Download of %s failed' % player_url)
820 res = self._parse_sig_js(code)
821 elif player_type == 'swf':
822 urlh = self._request_webpage(
823 player_url, video_id,
825 errnote='Download of %s failed' % player_url)
827 res = self._parse_sig_swf(code)
829 assert False, 'Invalid player type %r' % player_type
831 test_string = ''.join(map(compat_chr, range(len(example_sig))))
832 cache_res = res(test_string)
833 cache_spec = [ord(c) for c in cache_res]
835 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
838 def _print_sig_code(self, func, example_sig):
839 def gen_sig_code(idxs):
840 def _genslice(start, end, step):
841 starts = '' if start == 0 else str(start)
842 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
843 steps = '' if step == 1 else (':%d' % step)
844 return 's[%s%s%s]' % (starts, ends, steps)
847 # Quelch pyflakes warnings - start will be set when step is set
848 start = '(Never used)'
849 for i, prev in zip(idxs[1:], idxs[:-1]):
853 yield _genslice(start, prev, step)
856 if i - prev in [-1, 1]:
865 yield _genslice(start, i, step)
867 test_string = ''.join(map(compat_chr, range(len(example_sig))))
868 cache_res = func(test_string)
869 cache_spec = [ord(c) for c in cache_res]
870 expr_code = ' + '.join(gen_sig_code(cache_spec))
871 signature_id_tuple = '(%s)' % (
872 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
873 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
874 ' return %s\n') % (signature_id_tuple, expr_code)
875 self.to_screen('Extracted signature function:\n' + code)
877 def _parse_sig_js(self, jscode):
878 funcname = self._search_regex(
879 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
880 'Initial JS player signature function name')
882 jsi = JSInterpreter(jscode)
883 initial_function = jsi.extract_function(funcname)
884 return lambda s: initial_function([s])
886 def _parse_sig_swf(self, file_contents):
887 swfi = SWFInterpreter(file_contents)
888 TARGET_CLASSNAME = 'SignatureDecipher'
889 searched_class = swfi.extract_class(TARGET_CLASSNAME)
890 initial_function = swfi.extract_function(searched_class, 'decipher')
891 return lambda s: initial_function([s])
893 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
894 """Turn the encrypted s field into a working signature"""
896 if player_url is None:
897 raise ExtractorError('Cannot decrypt signature without player_url')
899 if player_url.startswith('//'):
900 player_url = 'https:' + player_url
902 player_id = (player_url, self._signature_cache_id(s))
903 if player_id not in self._player_cache:
904 func = self._extract_signature_function(
905 video_id, player_url, s
907 self._player_cache[player_id] = func
908 func = self._player_cache[player_id]
909 if self._downloader.params.get('youtube_print_sig_code'):
910 self._print_sig_code(func, s)
912 except Exception as e:
913 tb = traceback.format_exc()
914 raise ExtractorError(
915 'Signature extraction failed: ' + tb, cause=e)
917 def _get_subtitles(self, video_id, webpage):
919 subs_doc = self._download_xml(
920 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
921 video_id, note=False)
922 except ExtractorError as err:
923 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
927 for track in subs_doc.findall('track'):
928 lang = track.attrib['lang_code']
929 if lang in sub_lang_list:
932 for ext in self._SUBTITLE_FORMATS:
933 params = compat_urllib_parse.urlencode({
937 'name': track.attrib['name'].encode('utf-8'),
940 'url': 'https://www.youtube.com/api/timedtext?' + params,
943 sub_lang_list[lang] = sub_formats
944 if not sub_lang_list:
945 self._downloader.report_warning('video doesn\'t have subtitles')
949 def _get_ytplayer_config(self, video_id, webpage):
951 # User data may contain arbitrary character sequences that may affect
952 # JSON extraction with regex, e.g. when '};' is contained the second
953 # regex won't capture the whole JSON. Yet working around by trying more
954 # concrete regex first keeping in mind proper quoted string handling
955 # to be implemented in future that will replace this workaround (see
956 # https://github.com/rg3/youtube-dl/issues/7468,
957 # https://github.com/rg3/youtube-dl/pull/7599)
958 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
959 r';ytplayer\.config\s*=\s*({.+?});',
961 config = self._search_regex(
962 patterns, webpage, 'ytplayer.config', default=None)
964 return self._parse_json(
965 uppercase_escape(config), video_id, fatal=False)
967 def _get_automatic_captions(self, video_id, webpage):
968 """We need the webpage for getting the captions url, pass it as an
969 argument to speed up the process."""
970 self.to_screen('%s: Looking for automatic captions' % video_id)
971 player_config = self._get_ytplayer_config(video_id, webpage)
972 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
973 if not player_config:
974 self._downloader.report_warning(err_msg)
977 args = player_config['args']
978 caption_url = args['ttsurl']
980 self._downloader.report_warning(err_msg)
982 timestamp = args['timestamp']
983 # We get the available subtitles
984 list_params = compat_urllib_parse.urlencode({
989 list_url = caption_url + '&' + list_params
990 caption_list = self._download_xml(list_url, video_id)
991 original_lang_node = caption_list.find('track')
992 if original_lang_node is None:
993 self._downloader.report_warning('Video doesn\'t have automatic captions')
995 original_lang = original_lang_node.attrib['lang_code']
996 caption_kind = original_lang_node.attrib.get('kind', '')
999 for lang_node in caption_list.findall('target'):
1000 sub_lang = lang_node.attrib['lang_code']
1002 for ext in self._SUBTITLE_FORMATS:
1003 params = compat_urllib_parse.urlencode({
1004 'lang': original_lang,
1008 'kind': caption_kind,
1010 sub_formats.append({
1011 'url': caption_url + '&' + params,
1014 sub_lang_list[sub_lang] = sub_formats
1015 return sub_lang_list
1016 # An extractor error can be raise by the download process if there are
1017 # no automatic captions but there are subtitles
1018 except (KeyError, ExtractorError):
1019 self._downloader.report_warning(err_msg)
1023 def extract_id(cls, url):
1024 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1026 raise ExtractorError('Invalid URL: %s' % url)
1027 video_id = mobj.group(2)
1030 def _extract_from_m3u8(self, manifest_url, video_id):
1033 def _get_urls(_manifest):
1034 lines = _manifest.split('\n')
1035 urls = filter(lambda l: l and not l.startswith('#'),
1038 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1039 formats_urls = _get_urls(manifest)
1040 for format_url in formats_urls:
1041 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1042 url_map[itag] = format_url
1045 def _extract_annotations(self, video_id):
1046 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1047 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1049 def _real_extract(self, url):
1050 url, smuggled_data = unsmuggle_url(url, {})
1053 'http' if self._downloader.params.get('prefer_insecure', False)
1058 parsed_url = compat_urllib_parse_urlparse(url)
1059 for component in [parsed_url.fragment, parsed_url.query]:
1060 query = compat_parse_qs(component)
1061 if start_time is None and 't' in query:
1062 start_time = parse_duration(query['t'][0])
1063 if start_time is None and 'start' in query:
1064 start_time = parse_duration(query['start'][0])
1065 if end_time is None and 'end' in query:
1066 end_time = parse_duration(query['end'][0])
1068 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1069 mobj = re.search(self._NEXT_URL_RE, url)
1071 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1072 video_id = self.extract_id(url)
1075 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1076 video_webpage = self._download_webpage(url, video_id)
1078 # Attempt to extract SWF player URL
1079 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1080 if mobj is not None:
1081 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1087 def add_dash_mpd(video_info):
1088 dash_mpd = video_info.get('dashmpd')
1089 if dash_mpd and dash_mpd[0] not in dash_mpds:
1090 dash_mpds.append(dash_mpd[0])
1093 embed_webpage = None
1095 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1097 # We simulate the access to the video from www.youtube.com/v/{video_id}
1098 # this can be viewed without login into Youtube
1099 url = proto + '://www.youtube.com/embed/%s' % video_id
1100 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1101 data = compat_urllib_parse.urlencode({
1102 'video_id': video_id,
1103 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1104 'sts': self._search_regex(
1105 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1107 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1108 video_info_webpage = self._download_webpage(
1109 video_info_url, video_id,
1110 note='Refetching age-gated info webpage',
1111 errnote='unable to download video info webpage')
1112 video_info = compat_parse_qs(video_info_webpage)
1113 add_dash_mpd(video_info)
1117 # Try looking directly into the video webpage
1118 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1120 args = ytplayer_config['args']
1121 if args.get('url_encoded_fmt_stream_map'):
1122 # Convert to the same format returned by compat_parse_qs
1123 video_info = dict((k, [v]) for k, v in args.items())
1124 add_dash_mpd(video_info)
1125 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1127 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1128 # We also try looking in get_video_info since it may contain different dashmpd
1129 # URL that points to a DASH manifest with possibly different itag set (some itags
1130 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1131 # manifest pointed by get_video_info's dashmpd).
1132 # The general idea is to take a union of itags of both DASH manifests (for example
1133 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1134 self.report_video_info_webpage_download(video_id)
1135 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1137 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1138 % (proto, video_id, el_type))
1139 video_info_webpage = self._download_webpage(
1141 video_id, note=False,
1142 errnote='unable to download video info webpage')
1143 get_video_info = compat_parse_qs(video_info_webpage)
1144 if get_video_info.get('use_cipher_signature') != ['True']:
1145 add_dash_mpd(get_video_info)
1147 video_info = get_video_info
1148 if 'token' in get_video_info:
1149 # Different get_video_info requests may report different results, e.g.
1150 # some may report video unavailability, but some may serve it without
1151 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1152 # the original webpage as well as el=info and el=embedded get_video_info
1153 # requests report video unavailability due to geo restriction while
1154 # el=detailpage succeeds and returns valid data). This is probably
1155 # due to YouTube measures against IP ranges of hosting providers.
1156 # Working around by preferring the first succeeded video_info containing
1157 # the token if no such video_info yet was found.
1158 if 'token' not in video_info:
1159 video_info = get_video_info
1161 if 'token' not in video_info:
1162 if 'reason' in video_info:
1163 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1164 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1166 raise ExtractorError('YouTube said: This video is available in %s only' % (
1167 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1169 raise ExtractorError(
1170 'YouTube said: %s' % video_info['reason'][0],
1171 expected=True, video_id=video_id)
1173 raise ExtractorError(
1174 '"token" parameter not in video info for unknown reason',
1178 if 'title' in video_info:
1179 video_title = video_info['title'][0]
1181 self._downloader.report_warning('Unable to extract video title')
1185 video_description = get_element_by_id("eow-description", video_webpage)
1186 if video_description:
1187 video_description = re.sub(r'''(?x)
1189 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1190 (?:title|href)="([^"]+)"\s+
1191 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1192 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1195 ''', r'\1', video_description)
1196 video_description = clean_html(video_description)
1198 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1200 video_description = unescapeHTML(fd_mobj.group(1))
1202 video_description = ''
1204 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1205 if not self._downloader.params.get('noplaylist'):
1208 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1209 for feed in multifeed_metadata_list.split(','):
1210 # Unquote should take place before split on comma (,) since textual
1211 # fields may contain comma as well (see
1212 # https://github.com/rg3/youtube-dl/issues/8536)
1213 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1215 '_type': 'url_transparent',
1216 'ie_key': 'Youtube',
1218 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1219 {'force_singlefeed': True}),
1220 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1222 feed_ids.append(feed_data['id'][0])
1224 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1225 % (', '.join(feed_ids), video_id))
1226 return self.playlist_result(entries, video_id, video_title, video_description)
1227 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1229 if 'view_count' in video_info:
1230 view_count = int(video_info['view_count'][0])
1234 # Check for "rental" videos
1235 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1236 raise ExtractorError('"rental" videos not supported')
1238 # Start extracting information
1239 self.report_information_extraction(video_id)
1242 if 'author' not in video_info:
1243 raise ExtractorError('Unable to extract uploader name')
1244 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1247 video_uploader_id = None
1248 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1249 if mobj is not None:
1250 video_uploader_id = mobj.group(1)
1252 self._downloader.report_warning('unable to extract uploader nickname')
1255 # We try first to get a high quality image:
1256 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1257 video_webpage, re.DOTALL)
1258 if m_thumb is not None:
1259 video_thumbnail = m_thumb.group(1)
1260 elif 'thumbnail_url' not in video_info:
1261 self._downloader.report_warning('unable to extract video thumbnail')
1262 video_thumbnail = None
1263 else: # don't panic if we can't find it
1264 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1267 upload_date = self._html_search_meta(
1268 'datePublished', video_webpage, 'upload date', default=None)
1270 upload_date = self._search_regex(
1271 [r'(?s)id="eow-date.*?>(.*?)</span>',
1272 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1273 video_webpage, 'upload date', default=None)
1275 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1276 upload_date = unified_strdate(upload_date)
1278 m_music = re.search(
1279 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1282 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1283 video_creator = clean_html(m_music.group('creator'))
1285 video_alt_title = video_creator = None
1287 m_cat_container = self._search_regex(
1288 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1289 video_webpage, 'categories', default=None)
1291 category = self._html_search_regex(
1292 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1294 video_categories = None if category is None else [category]
1296 video_categories = None
1299 unescapeHTML(m.group('content'))
1300 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1302 def _extract_count(count_name):
1303 return str_to_int(self._search_regex(
1304 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1305 % re.escape(count_name),
1306 video_webpage, count_name, default=None))
1308 like_count = _extract_count('like')
1309 dislike_count = _extract_count('dislike')
1312 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1313 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1315 if 'length_seconds' not in video_info:
1316 self._downloader.report_warning('unable to extract video duration')
1317 video_duration = None
1319 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1322 video_annotations = None
1323 if self._downloader.params.get('writeannotations', False):
1324 video_annotations = self._extract_annotations(video_id)
1326 def _map_to_format_list(urlmap):
1328 for itag, video_real_url in urlmap.items():
1331 'url': video_real_url,
1332 'player_url': player_url,
1334 if itag in self._formats:
1335 dct.update(self._formats[itag])
1339 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1340 self.report_rtmp_download()
1342 'format_id': '_rtmp',
1344 'url': video_info['conn'][0],
1345 'player_url': player_url,
1347 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1348 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1349 if 'rtmpe%3Dyes' in encoded_url_map:
1350 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1352 for url_data_str in encoded_url_map.split(','):
1353 url_data = compat_parse_qs(url_data_str)
1354 if 'itag' not in url_data or 'url' not in url_data:
1356 format_id = url_data['itag'][0]
1357 url = url_data['url'][0]
1359 if 'sig' in url_data:
1360 url += '&signature=' + url_data['sig'][0]
1361 elif 's' in url_data:
1362 encrypted_sig = url_data['s'][0]
1363 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1365 jsplayer_url_json = self._search_regex(
1367 embed_webpage if age_gate else video_webpage,
1368 'JS player URL (1)', default=None)
1369 if not jsplayer_url_json and not age_gate:
1370 # We need the embed website after all
1371 if embed_webpage is None:
1372 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1373 embed_webpage = self._download_webpage(
1374 embed_url, video_id, 'Downloading embed webpage')
1375 jsplayer_url_json = self._search_regex(
1376 ASSETS_RE, embed_webpage, 'JS player URL')
1378 player_url = json.loads(jsplayer_url_json)
1379 if player_url is None:
1380 player_url_json = self._search_regex(
1381 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1382 video_webpage, 'age gate player URL')
1383 player_url = json.loads(player_url_json)
1385 if self._downloader.params.get('verbose'):
1386 if player_url is None:
1387 player_version = 'unknown'
1388 player_desc = 'unknown'
1390 if player_url.endswith('swf'):
1391 player_version = self._search_regex(
1392 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1393 'flash player', fatal=False)
1394 player_desc = 'flash player %s' % player_version
1396 player_version = self._search_regex(
1397 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1399 'html5 player', fatal=False)
1400 player_desc = 'html5 player %s' % player_version
1402 parts_sizes = self._signature_cache_id(encrypted_sig)
1403 self.to_screen('{%s} signature length %s, %s' %
1404 (format_id, parts_sizes, player_desc))
1406 signature = self._decrypt_signature(
1407 encrypted_sig, video_id, player_url, age_gate)
1408 url += '&signature=' + signature
1409 if 'ratebypass' not in url:
1410 url += '&ratebypass=yes'
1413 'format_id': format_id,
1415 'player_url': player_url,
1417 if format_id in self._formats:
1418 dct.update(self._formats[format_id])
1420 # Some itags are not included in DASH manifest thus corresponding formats will
1421 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1422 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1423 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1424 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1427 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1428 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1431 'fps': int_or_none(url_data.get('fps', [None])[0]),
1432 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1434 for key, value in more_fields.items():
1437 type_ = url_data.get('type', [None])[0]
1439 type_split = type_.split(';')
1440 kind_ext = type_split[0].split('/')
1441 if len(kind_ext) == 2:
1443 dct['ext'] = mimetype2ext(type_split[0])
1444 if kind in ('audio', 'video'):
1446 for mobj in re.finditer(
1447 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1448 if mobj.group('key') == 'codecs':
1449 codecs = mobj.group('val')
1452 codecs = codecs.split(',')
1453 if len(codecs) == 2:
1454 acodec, vcodec = codecs[1], codecs[0]
1456 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1462 elif video_info.get('hlsvp'):
1463 manifest_url = video_info['hlsvp'][0]
1464 url_map = self._extract_from_m3u8(manifest_url, video_id)
1465 formats = _map_to_format_list(url_map)
1466 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1467 for a_format in formats:
1468 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1470 unavailable_message = self._html_search_regex(
1471 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1472 video_webpage, 'unavailable message', default=None)
1473 if unavailable_message:
1474 raise ExtractorError(unavailable_message, expected=True)
1475 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1477 # Look for the DASH manifest
1478 if self._downloader.params.get('youtube_include_dash_manifest', True):
1479 dash_mpd_fatal = True
1480 for mpd_url in dash_mpds:
1483 def decrypt_sig(mobj):
1485 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1486 return '/signature/%s' % dec_s
1488 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1490 for df in self._extract_mpd_formats(
1491 mpd_url, video_id, fatal=dash_mpd_fatal,
1492 formats_dict=self._formats):
1493 # Do not overwrite DASH format found in some previous DASH manifest
1494 if df['format_id'] not in dash_formats:
1495 dash_formats[df['format_id']] = df
1496 # Additional DASH manifests may end up in HTTP Error 403 therefore
1497 # allow them to fail without bug report message if we already have
1498 # some DASH manifest succeeded. This is temporary workaround to reduce
1499 # burst of bug reports until we figure out the reason and whether it
1500 # can be fixed at all.
1501 dash_mpd_fatal = False
1502 except (ExtractorError, KeyError) as e:
1503 self.report_warning(
1504 'Skipping DASH manifest: %r' % e, video_id)
1506 # Remove the formats we found through non-DASH, they
1507 # contain less info and it can be wrong, because we use
1508 # fixed values (for example the resolution). See
1509 # https://github.com/rg3/youtube-dl/issues/5774 for an
1511 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1512 formats.extend(dash_formats.values())
1514 # Check for malformed aspect ratio
1515 stretched_m = re.search(
1516 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1519 w = float(stretched_m.group('w'))
1520 h = float(stretched_m.group('h'))
1521 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1522 # We will only process correct ratios.
1526 if f.get('vcodec') != 'none':
1527 f['stretched_ratio'] = ratio
1529 self._sort_formats(formats)
1533 'uploader': video_uploader,
1534 'uploader_id': video_uploader_id,
1535 'upload_date': upload_date,
1536 'creator': video_creator,
1537 'title': video_title,
1538 'alt_title': video_alt_title,
1539 'thumbnail': video_thumbnail,
1540 'description': video_description,
1541 'categories': video_categories,
1543 'subtitles': video_subtitles,
1544 'automatic_captions': automatic_captions,
1545 'duration': video_duration,
1546 'age_limit': 18 if age_gate else 0,
1547 'annotations': video_annotations,
1548 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1549 'view_count': view_count,
1550 'like_count': like_count,
1551 'dislike_count': dislike_count,
1552 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1555 'start_time': start_time,
1556 'end_time': end_time,
1560 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1561 IE_DESC = 'YouTube.com playlists'
1562 _VALID_URL = r"""(?x)(?:
1567 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1568 \? (?:.*?[&;])*? (?:p|a|list)=
1572 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1573 # Top tracks, they can also include dots
1578 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1580 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1581 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1582 IE_NAME = 'youtube:playlist'
1584 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1586 'title': 'ytdl test PL',
1587 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1589 'playlist_count': 3,
1591 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1593 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1594 'title': 'YDL_Empty_List',
1596 'playlist_count': 0,
1598 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1599 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1601 'title': '29C3: Not my department',
1602 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1604 'playlist_count': 95,
1606 'note': 'issue #673',
1607 'url': 'PLBB231211A4F62143',
1609 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1610 'id': 'PLBB231211A4F62143',
1612 'playlist_mincount': 26,
1614 'note': 'Large playlist',
1615 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1617 'title': 'Uploads from Cauchemar',
1618 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1620 'playlist_mincount': 799,
1622 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1624 'title': 'YDL_safe_search',
1625 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1627 'playlist_count': 2,
1630 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1631 'playlist_count': 4,
1634 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1637 'note': 'Embedded SWF player',
1638 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1639 'playlist_count': 4,
1642 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1645 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1646 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1648 'title': 'Uploads from Interstellar Movie',
1649 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1651 'playlist_mincout': 21,
1654 def _real_initialize(self):
1657 def _extract_mix(self, playlist_id):
1658 # The mixes are generated from a single video
1659 # the id of the playlist is just 'RD' + video_id
1660 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1661 webpage = self._download_webpage(
1662 url, playlist_id, 'Downloading Youtube mix')
1663 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1665 search_title('playlist-title') or
1666 search_title('title long-title') or
1667 search_title('title'))
1668 title = clean_html(title_span)
1669 ids = orderedSet(re.findall(
1670 r'''(?xs)data-video-username=".*?".*?
1671 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1673 url_results = self._ids_to_results(ids)
1675 return self.playlist_result(url_results, playlist_id, title)
1677 def _extract_playlist(self, playlist_id):
1678 url = self._TEMPLATE_URL % playlist_id
1679 page = self._download_webpage(url, playlist_id)
1681 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1682 match = match.strip()
1683 # Check if the playlist exists or is private
1684 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1685 raise ExtractorError(
1686 'The playlist doesn\'t exist or is private, use --username or '
1687 '--netrc to access it.',
1689 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1690 raise ExtractorError(
1691 'Invalid parameters. Maybe URL is incorrect.',
1693 elif re.match(r'[^<]*Choose your language[^<]*', match):
1696 self.report_warning('Youtube gives an alert message: ' + match)
1698 playlist_title = self._html_search_regex(
1699 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1702 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1704 def _check_download_just_video(self, url, playlist_id):
1705 # Check if it's a video-specific URL
1706 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1707 if 'v' in query_dict:
1708 video_id = query_dict['v'][0]
1709 if self._downloader.params.get('noplaylist'):
1710 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1711 return self.url_result(video_id, 'Youtube', video_id=video_id)
1713 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1715 def _real_extract(self, url):
1716 # Extract playlist id
1717 mobj = re.match(self._VALID_URL, url)
1719 raise ExtractorError('Invalid URL: %s' % url)
1720 playlist_id = mobj.group(1) or mobj.group(2)
1722 video = self._check_download_just_video(url, playlist_id)
1726 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1727 # Mixes require a custom extraction process
1728 return self._extract_mix(playlist_id)
1730 return self._extract_playlist(playlist_id)
1733 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1734 IE_DESC = 'YouTube.com channels'
1735 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1736 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1737 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1738 IE_NAME = 'youtube:channel'
1740 'note': 'paginated channel',
1741 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1742 'playlist_mincount': 91,
1744 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1745 'title': 'Uploads from lex will',
1748 'note': 'Age restricted channel',
1749 # from https://www.youtube.com/user/DeusExOfficial
1750 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1751 'playlist_mincount': 64,
1753 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1754 'title': 'Uploads from Deus Ex',
1759 def suitable(cls, url):
1760 return False if YoutubePlaylistsIE.suitable(url) else super(YoutubeChannelIE, cls).suitable(url)
1762 def _real_extract(self, url):
1763 channel_id = self._match_id(url)
1765 url = self._TEMPLATE_URL % channel_id
1767 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1768 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1769 # otherwise fallback on channel by page extraction
1770 channel_page = self._download_webpage(
1771 url + '?view=57', channel_id,
1772 'Downloading channel page', fatal=False)
1773 if channel_page is False:
1774 channel_playlist_id = False
1776 channel_playlist_id = self._html_search_meta(
1777 'channelId', channel_page, 'channel id', default=None)
1778 if not channel_playlist_id:
1779 channel_playlist_id = self._search_regex(
1780 r'data-(?:channel-external-|yt)id="([^"]+)"',
1781 channel_page, 'channel id', default=None)
1782 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1783 playlist_id = 'UU' + channel_playlist_id[2:]
1784 return self.url_result(
1785 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1787 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1788 autogenerated = re.search(r'''(?x)
1790 channel-header-autogenerated-label|
1791 yt-channel-title-autogenerated
1792 )[^"]*"''', channel_page) is not None
1795 # The videos are contained in a single page
1796 # the ajax pages can't be used, they are empty
1799 video_id, 'Youtube', video_id=video_id,
1800 video_title=video_title)
1801 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1802 return self.playlist_result(entries, channel_id)
1804 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1807 class YoutubeUserIE(YoutubeChannelIE):
1808 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1809 _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_-]+)'
1810 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1811 IE_NAME = 'youtube:user'
1814 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1815 'playlist_mincount': 320,
1817 'title': 'TheLinuxFoundation',
1820 'url': 'ytuser:phihag',
1821 'only_matching': True,
1825 def suitable(cls, url):
1826 # Don't return True if the url can be extracted with other youtube
1827 # extractor, the regex would is too permissive and it would match.
1828 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1829 if any(ie.suitable(url) for ie in other_ies):
1832 return super(YoutubeUserIE, cls).suitable(url)
1835 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1836 IE_DESC = 'YouTube.com user/channel playlists'
1837 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
1838 IE_NAME = 'youtube:playlists'
1841 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1842 'playlist_mincount': 4,
1844 'id': 'ThirstForScience',
1845 'title': 'Thirst for Science',
1848 # with "Load more" button
1849 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1850 'playlist_mincount': 70,
1853 'title': 'Игорь Клейнер',
1856 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
1857 'playlist_mincount': 17,
1859 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
1860 'title': 'Chem Player',
1865 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1866 IE_DESC = 'YouTube.com searches'
1867 # there doesn't appear to be a real limit, for example if you search for
1868 # 'python' you get more than 8.000.000 results
1869 _MAX_RESULTS = float('inf')
1870 IE_NAME = 'youtube:search'
1871 _SEARCH_KEY = 'ytsearch'
1872 _EXTRA_QUERY_ARGS = {}
1875 def _get_n_results(self, query, n):
1876 """Get a specified number of results for a query"""
1881 for pagenum in itertools.count(1):
1883 'search_query': query.encode('utf-8'),
1887 url_query.update(self._EXTRA_QUERY_ARGS)
1888 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1889 data = self._download_json(
1890 result_url, video_id='query "%s"' % query,
1891 note='Downloading page %s' % pagenum,
1892 errnote='Unable to download API page')
1893 html_content = data[1]['body']['content']
1895 if 'class="search-message' in html_content:
1896 raise ExtractorError(
1897 '[youtube] No video results', expected=True)
1899 new_videos = self._ids_to_results(orderedSet(re.findall(
1900 r'href="/watch\?v=(.{11})', html_content)))
1901 videos += new_videos
1902 if not new_videos or len(videos) > limit:
1907 return self.playlist_result(videos, query)
1910 class YoutubeSearchDateIE(YoutubeSearchIE):
1911 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1912 _SEARCH_KEY = 'ytsearchdate'
1913 IE_DESC = 'YouTube.com searches, newest videos first'
1914 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1917 class YoutubeSearchURLIE(InfoExtractor):
1918 IE_DESC = 'YouTube.com search URLs'
1919 IE_NAME = 'youtube:search_url'
1920 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
1922 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1923 'playlist_mincount': 5,
1925 'title': 'youtube-dl test video',
1928 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
1929 'only_matching': True,
1932 def _real_extract(self, url):
1933 mobj = re.match(self._VALID_URL, url)
1934 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
1936 webpage = self._download_webpage(url, query)
1937 result_code = self._search_regex(
1938 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1940 part_codes = re.findall(
1941 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
1943 for part_code in part_codes:
1944 part_title = self._html_search_regex(
1945 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1946 part_url_snippet = self._html_search_regex(
1947 r'(?s)href="([^"]+)"', part_code, 'item URL')
1948 part_url = compat_urlparse.urljoin(
1949 'https://www.youtube.com/', part_url_snippet)
1953 'title': part_title,
1957 '_type': 'playlist',
1963 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
1964 IE_DESC = 'YouTube.com (multi-season) shows'
1965 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1966 IE_NAME = 'youtube:show'
1968 'url': 'https://www.youtube.com/show/airdisasters',
1969 'playlist_mincount': 5,
1971 'id': 'airdisasters',
1972 'title': 'Air Disasters',
1976 def _real_extract(self, url):
1977 playlist_id = self._match_id(url)
1978 return super(YoutubeShowIE, self)._real_extract(
1979 'https://www.youtube.com/show/%s/playlists' % playlist_id)
1982 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1984 Base class for feed extractors
1985 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1987 _LOGIN_REQUIRED = True
1991 return 'youtube:%s' % self._FEED_NAME
1993 def _real_initialize(self):
1996 def _real_extract(self, url):
1997 page = self._download_webpage(
1998 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2000 # The extraction process is the same as for playlists, but the regex
2001 # for the video ids doesn't contain an index
2003 more_widget_html = content_html = page
2004 for page_num in itertools.count(1):
2005 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2007 # 'recommended' feed has infinite 'load more' and each new portion spins
2008 # the same videos in (sometimes) slightly different order, so we'll check
2009 # for unicity and break when portion has no new videos
2010 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2016 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2020 more = self._download_json(
2021 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2022 'Downloading page #%s' % page_num,
2023 transform_source=uppercase_escape)
2024 content_html = more['content_html']
2025 more_widget_html = more['load_more_widget_html']
2027 return self.playlist_result(
2028 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2031 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2032 IE_NAME = 'youtube:watchlater'
2033 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2034 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2037 'url': 'https://www.youtube.com/playlist?list=WL',
2038 'only_matching': True,
2040 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2041 'only_matching': True,
2044 def _real_extract(self, url):
2045 video = self._check_download_just_video(url, 'WL')
2048 return self._extract_playlist('WL')
2051 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2052 IE_NAME = 'youtube:favorites'
2053 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2054 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2055 _LOGIN_REQUIRED = True
2057 def _real_extract(self, url):
2058 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2059 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2060 return self.url_result(playlist_id, 'YoutubePlaylist')
2063 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2064 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2065 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2066 _FEED_NAME = 'recommended'
2067 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2070 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2071 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2072 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2073 _FEED_NAME = 'subscriptions'
2074 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2077 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2078 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2079 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2080 _FEED_NAME = 'history'
2081 _PLAYLIST_TITLE = 'Youtube History'
2084 class YoutubeTruncatedURLIE(InfoExtractor):
2085 IE_NAME = 'youtube:truncated_url'
2086 IE_DESC = False # Do not list
2087 _VALID_URL = r'''(?x)
2089 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2092 annotation_id=annotation_[^&]+|
2098 attribution_link\?a=[^&]+
2104 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2105 'only_matching': True,
2107 'url': 'http://www.youtube.com/watch?',
2108 'only_matching': True,
2110 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2111 'only_matching': True,
2113 'url': 'https://www.youtube.com/watch?feature=foo',
2114 'only_matching': True,
2116 'url': 'https://www.youtube.com/watch?hl=en-GB',
2117 'only_matching': True,
2119 'url': 'https://www.youtube.com/watch?t=2372',
2120 'only_matching': True,
2123 def _real_extract(self, url):
2124 raise ExtractorError(
2125 'Did you forget to quote the URL? Remember that & is a meta '
2126 'character in most shells, so you want to put the URL in quotes, '
2128 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2129 ' or simply youtube-dl BaW_jenozKc .',
2133 class YoutubeTruncatedIDIE(InfoExtractor):
2134 IE_NAME = 'youtube:truncated_id'
2135 IE_DESC = False # Do not list
2136 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2139 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2140 'only_matching': True,
2143 def _real_extract(self, url):
2144 video_id = self._match_id(url)
2145 raise ExtractorError(
2146 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),