3 from __future__ import unicode_literals
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
21 compat_urllib_parse_unquote,
22 compat_urllib_parse_unquote_plus,
23 compat_urllib_parse_urlparse,
33 get_element_by_attribute,
52 class YoutubeBaseInfoExtractor(InfoExtractor):
53 """Provide base functions for Youtube extractors"""
54 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
55 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
56 _NETRC_MACHINE = 'youtube'
57 # If True it will raise an error if no login info is provided
58 _LOGIN_REQUIRED = False
60 def _set_language(self):
62 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
63 # YouTube sets the expire time to about two months
64 expire_time=time.time() + 2 * 30 * 24 * 3600)
66 def _ids_to_results(self, ids):
68 self.url_result(vid_id, 'Youtube', video_id=vid_id)
73 Attempt to log in to YouTube.
74 True is returned if successful or skipped.
75 False is returned if login failed.
77 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
79 (username, password) = self._get_login_info()
80 # No authentication to be performed
82 if self._LOGIN_REQUIRED:
83 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
86 login_page = self._download_webpage(
87 self._LOGIN_URL, None,
88 note='Downloading login page',
89 errnote='unable to fetch login page', fatal=False)
90 if login_page is False:
93 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
94 login_page, 'Login GALX parameter')
98 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
103 'PersistentCookie': 'yes',
105 'bgresponse': 'js_disabled',
106 'checkConnection': '',
107 'checkedDomains': 'youtube',
114 'service': 'youtube',
119 login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii')
121 req = sanitized_Request(self._LOGIN_URL, login_data)
122 login_results = self._download_webpage(
124 note='Logging in', errnote='unable to log in', fatal=False)
125 if login_results is False:
128 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
129 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
132 # TODO add SMS and phone call support - these require making a request and then prompting the user
134 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
135 tfa_code = self._get_tfa_info('2-step verification code')
138 self._downloader.report_warning(
139 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
140 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
143 tfa_code = remove_start(tfa_code, 'G-')
145 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
147 tfa_form_strs.update({
152 tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii')
154 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
155 tfa_results = self._download_webpage(
157 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
159 if tfa_results is False:
162 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
163 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
165 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
166 self._downloader.report_warning('unable to log in - did the page structure change?')
168 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
169 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
172 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
173 self._downloader.report_warning('unable to log in: bad username or password')
177 def _real_initialize(self):
178 if self._downloader is None:
181 if not self._login():
185 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
186 # Extract entries from page with "Load more" button
187 def _entries(self, page, playlist_id):
188 more_widget_html = content_html = page
189 for page_num in itertools.count(1):
190 for entry in self._process_page(content_html):
193 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
197 more = self._download_json(
198 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
199 'Downloading page #%s' % page_num,
200 transform_source=uppercase_escape)
201 content_html = more['content_html']
202 if not content_html.strip():
203 # Some webpages show a "Load more" button but they don't
206 more_widget_html = more['load_more_widget_html']
209 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
210 def _process_page(self, content):
211 for video_id, video_title in self.extract_videos_from_page(content):
212 yield self.url_result(video_id, 'Youtube', video_id, video_title)
214 def extract_videos_from_page(self, page):
217 for mobj in re.finditer(self._VIDEO_RE, page):
218 # The link with index 0 is not the first video of the playlist (not sure if still actual)
219 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
221 video_id = mobj.group('id')
222 video_title = unescapeHTML(mobj.group('title'))
224 video_title = video_title.strip()
226 idx = ids_in_page.index(video_id)
227 if video_title and not titles_in_page[idx]:
228 titles_in_page[idx] = video_title
230 ids_in_page.append(video_id)
231 titles_in_page.append(video_title)
232 return zip(ids_in_page, titles_in_page)
235 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
236 def _process_page(self, content):
237 for playlist_id in orderedSet(re.findall(r'href="/?playlist\?list=([0-9A-Za-z-_]{10,})"', content)):
238 yield self.url_result(
239 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
241 def _real_extract(self, url):
242 playlist_id = self._match_id(url)
243 webpage = self._download_webpage(url, playlist_id)
244 title = self._og_search_title(webpage, fatal=False)
245 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
248 class YoutubeIE(YoutubeBaseInfoExtractor):
249 IE_DESC = 'YouTube.com'
250 _VALID_URL = r"""(?x)^
252 (?:https?://|//) # http(s):// or protocol-independent URL
253 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
254 (?:www\.)?deturl\.com/www\.youtube\.com/|
255 (?:www\.)?pwnyoutube\.com/|
256 (?:www\.)?yourepeat\.com/|
257 tube\.majestyc\.net/|
258 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
259 (?:.*?\#/)? # handle anchor (#/) redirect urls
260 (?: # the various things that can precede the ID:
261 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
262 |(?: # or the v= param in all its forms
263 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
264 (?:\?|\#!?) # the params delimiter ? or # or #!
265 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
270 youtu\.be| # just youtu.be/xxxx
271 vid\.plus # or vid.plus/xxxx
273 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
275 )? # all until now is optional -> you can pass the naked ID
276 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
277 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
278 (?(1).+)? # if we found the ID, everything can follow
280 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
282 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
283 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
284 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
285 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
286 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
287 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
288 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
289 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
290 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
291 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
292 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
293 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
294 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
295 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
296 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
297 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
298 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
299 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
303 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
304 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
305 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
306 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
307 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
308 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
309 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
311 # Apple HTTP Live Streaming
312 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
313 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
314 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
315 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
316 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
317 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
318 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
321 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
322 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
323 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
324 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
325 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
326 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
327 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
328 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
329 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
330 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
331 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
334 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
335 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
336 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
339 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
340 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
341 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
342 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
343 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
344 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
345 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
346 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
347 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
348 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
349 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
350 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
351 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
352 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
353 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
354 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
355 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
356 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
357 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
358 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
359 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
360 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
363 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
364 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
366 # Dash webm audio with opus inside
367 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
368 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
369 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
372 '_rtmp': {'protocol': 'rtmp'},
374 _SUBTITLE_FORMATS = ('ttml', 'vtt')
379 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
383 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
384 'uploader': 'Philipp Hagemeister',
385 'uploader_id': 'phihag',
386 'upload_date': '20121002',
387 '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 .',
388 'categories': ['Science & Technology'],
389 'tags': ['youtube-dl'],
391 'dislike_count': int,
397 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
398 'note': 'Test generic use_cipher_signature video (#897)',
402 'upload_date': '20120506',
403 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
404 'alt_title': 'I Love It (feat. Charli XCX)',
405 'description': 'md5:782e8651347686cba06e58f71ab51773',
406 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
407 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
408 'iconic ep', 'iconic', 'love', 'it'],
409 'uploader': 'Icona Pop',
410 'uploader_id': 'IconaPop',
411 'creator': 'Icona Pop',
415 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
416 'note': 'Test VEVO video with age protection (#956)',
420 'upload_date': '20130703',
421 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
422 'alt_title': 'Tunnel Vision',
423 'description': 'md5:64249768eec3bc4276236606ea996373',
424 'uploader': 'justintimberlakeVEVO',
425 'uploader_id': 'justintimberlakeVEVO',
426 'creator': 'Justin Timberlake',
431 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
432 'note': 'Embed-only video (#1746)',
436 'upload_date': '20120608',
437 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
438 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
439 'uploader': 'SET India',
440 'uploader_id': 'setindia',
445 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
446 'note': 'Use the first video ID in the URL',
450 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
451 'uploader': 'Philipp Hagemeister',
452 'uploader_id': 'phihag',
453 'upload_date': '20121002',
454 '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 .',
455 'categories': ['Science & Technology'],
456 'tags': ['youtube-dl'],
458 'dislike_count': int,
461 'skip_download': True,
465 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
466 'note': '256k DASH audio (format 141) via DASH manifest',
470 'upload_date': '20121002',
471 'uploader_id': '8KVIDEO',
473 'uploader': '8KVIDEO',
474 'title': 'UHDTV TEST 8K VIDEO.mp4'
477 'youtube_include_dash_manifest': True,
481 # DASH manifest with encrypted signature
483 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
487 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
488 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
489 'uploader': 'AfrojackVEVO',
490 'uploader_id': 'AfrojackVEVO',
491 'upload_date': '20131011',
494 'youtube_include_dash_manifest': True,
498 # JS player signature function name containing $
500 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
504 'title': 'Taylor Swift - Shake It Off',
505 'alt_title': 'Shake It Off',
506 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
507 'uploader': 'TaylorSwiftVEVO',
508 'uploader_id': 'TaylorSwiftVEVO',
509 'upload_date': '20140818',
510 'creator': 'Taylor Swift',
513 'youtube_include_dash_manifest': True,
519 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
523 'upload_date': '20100909',
524 'uploader': 'The Amazing Atheist',
525 'uploader_id': 'TheAmazingAtheist',
526 'title': 'Burning Everyone\'s Koran',
527 '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',
530 # Normal age-gate video (No vevo, embed allowed)
532 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
536 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
537 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
538 'uploader': 'The Witcher',
539 'uploader_id': 'WitcherGame',
540 'upload_date': '20140605',
544 # Age-gate video with encrypted signature
546 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
550 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
551 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
552 'uploader': 'LloydVEVO',
553 'uploader_id': 'LloydVEVO',
554 'upload_date': '20110629',
558 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
560 'url': '__2ABJjxzNo',
564 'upload_date': '20100430',
565 'uploader_id': 'deadmau5',
566 'creator': 'deadmau5',
567 'description': 'md5:12c56784b8032162bb936a5f76d55360',
568 'uploader': 'deadmau5',
569 'title': 'Deadmau5 - Some Chords (HD)',
570 'alt_title': 'Some Chords',
572 'expected_warnings': [
573 'DASH manifest missing',
576 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
578 'url': 'lqQg6PlCWgI',
582 'upload_date': '20150827',
583 'uploader_id': 'olympic',
584 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
585 'uploader': 'Olympics',
586 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
589 'skip_download': 'requires avconv',
594 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
598 'stretched_ratio': 16 / 9.,
599 'upload_date': '20110310',
600 'uploader_id': 'AllenMeow',
601 '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',
603 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
606 # url_encoded_fmt_stream_map is empty string
608 'url': 'qEJwOuvDf7I',
612 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
614 'upload_date': '20150404',
615 'uploader_id': 'spbelect',
616 'uploader': 'Наблюдатели Петербурга',
619 'skip_download': 'requires avconv',
621 'skip': 'This live event has ended.',
623 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
625 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
629 'title': 'md5:7b81415841e02ecd4313668cde88737a',
630 'description': 'md5:116377fd2963b81ec4ce64b542173306',
631 'upload_date': '20150625',
632 'uploader_id': 'dorappi2000',
633 'uploader': 'dorappi2000',
634 'formats': 'mincount:33',
637 # DASH manifest with segment_list
639 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
640 'md5': '8ce563a1d667b599d21064e982ab9e31',
644 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
645 'uploader': 'Airtek',
646 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
647 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
648 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
651 'youtube_include_dash_manifest': True,
652 'format': '135', # bestvideo
656 # Multifeed videos (multiple cameras), URL is for Main Camera
657 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
660 'title': 'teamPGP: Rocket League Noob Stream',
661 'description': 'md5:dc7872fb300e143831327f1bae3af010',
667 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
668 'description': 'md5:dc7872fb300e143831327f1bae3af010',
669 'upload_date': '20150721',
670 'uploader': 'Beer Games Beer',
671 'uploader_id': 'beergamesbeer',
677 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
678 'description': 'md5:dc7872fb300e143831327f1bae3af010',
679 'upload_date': '20150721',
680 'uploader': 'Beer Games Beer',
681 'uploader_id': 'beergamesbeer',
687 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
688 'description': 'md5:dc7872fb300e143831327f1bae3af010',
689 'upload_date': '20150721',
690 'uploader': 'Beer Games Beer',
691 'uploader_id': 'beergamesbeer',
697 'title': 'teamPGP: Rocket League Noob Stream (zim)',
698 'description': 'md5:dc7872fb300e143831327f1bae3af010',
699 'upload_date': '20150721',
700 'uploader': 'Beer Games Beer',
701 'uploader_id': 'beergamesbeer',
705 'skip_download': True,
709 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
710 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
713 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
718 'url': 'http://vid.plus/FlRa-iH7PGw',
719 'only_matching': True,
722 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
723 # Also tests cut-off URL expansion in video description (see
724 # https://github.com/rg3/youtube-dl/issues/1892,
725 # https://github.com/rg3/youtube-dl/issues/8164)
726 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
730 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
731 'alt_title': 'Dark Walk',
732 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
733 'upload_date': '20151119',
734 'uploader_id': 'IronSoulElf',
735 'uploader': 'IronSoulElf',
736 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
739 'skip_download': True,
743 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
744 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
745 'only_matching': True,
748 # Video with yt:stretch=17:0
749 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
753 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
754 'description': 'md5:ee18a25c350637c8faff806845bddee9',
755 'upload_date': '20151107',
756 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
757 'uploader': 'CH GAMER DROID',
760 'skip_download': True,
764 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
765 'only_matching': True,
769 def __init__(self, *args, **kwargs):
770 super(YoutubeIE, self).__init__(*args, **kwargs)
771 self._player_cache = {}
773 def report_video_info_webpage_download(self, video_id):
774 """Report attempt to download video info webpage."""
775 self.to_screen('%s: Downloading video info webpage' % video_id)
777 def report_information_extraction(self, video_id):
778 """Report attempt to extract video information."""
779 self.to_screen('%s: Extracting video information' % video_id)
781 def report_unavailable_format(self, video_id, format):
782 """Report extracted video URL."""
783 self.to_screen('%s: Format %s not available' % (video_id, format))
785 def report_rtmp_download(self):
786 """Indicate the download will use the RTMP protocol."""
787 self.to_screen('RTMP download detected')
789 def _signature_cache_id(self, example_sig):
790 """ Return a string representation of a signature """
791 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
793 def _extract_signature_function(self, video_id, player_url, example_sig):
795 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
798 raise ExtractorError('Cannot identify player %r' % player_url)
799 player_type = id_m.group('ext')
800 player_id = id_m.group('id')
802 # Read from filesystem cache
803 func_id = '%s_%s_%s' % (
804 player_type, player_id, self._signature_cache_id(example_sig))
805 assert os.path.basename(func_id) == func_id
807 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
808 if cache_spec is not None:
809 return lambda s: ''.join(s[i] for i in cache_spec)
812 'Downloading player %s' % player_url
813 if self._downloader.params.get('verbose') else
814 'Downloading %s player %s' % (player_type, player_id)
816 if player_type == 'js':
817 code = self._download_webpage(
818 player_url, video_id,
820 errnote='Download of %s failed' % player_url)
821 res = self._parse_sig_js(code)
822 elif player_type == 'swf':
823 urlh = self._request_webpage(
824 player_url, video_id,
826 errnote='Download of %s failed' % player_url)
828 res = self._parse_sig_swf(code)
830 assert False, 'Invalid player type %r' % player_type
832 test_string = ''.join(map(compat_chr, range(len(example_sig))))
833 cache_res = res(test_string)
834 cache_spec = [ord(c) for c in cache_res]
836 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
839 def _print_sig_code(self, func, example_sig):
840 def gen_sig_code(idxs):
841 def _genslice(start, end, step):
842 starts = '' if start == 0 else str(start)
843 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
844 steps = '' if step == 1 else (':%d' % step)
845 return 's[%s%s%s]' % (starts, ends, steps)
848 # Quelch pyflakes warnings - start will be set when step is set
849 start = '(Never used)'
850 for i, prev in zip(idxs[1:], idxs[:-1]):
854 yield _genslice(start, prev, step)
857 if i - prev in [-1, 1]:
866 yield _genslice(start, i, step)
868 test_string = ''.join(map(compat_chr, range(len(example_sig))))
869 cache_res = func(test_string)
870 cache_spec = [ord(c) for c in cache_res]
871 expr_code = ' + '.join(gen_sig_code(cache_spec))
872 signature_id_tuple = '(%s)' % (
873 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
874 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
875 ' return %s\n') % (signature_id_tuple, expr_code)
876 self.to_screen('Extracted signature function:\n' + code)
878 def _parse_sig_js(self, jscode):
879 funcname = self._search_regex(
880 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
881 'Initial JS player signature function name')
883 jsi = JSInterpreter(jscode)
884 initial_function = jsi.extract_function(funcname)
885 return lambda s: initial_function([s])
887 def _parse_sig_swf(self, file_contents):
888 swfi = SWFInterpreter(file_contents)
889 TARGET_CLASSNAME = 'SignatureDecipher'
890 searched_class = swfi.extract_class(TARGET_CLASSNAME)
891 initial_function = swfi.extract_function(searched_class, 'decipher')
892 return lambda s: initial_function([s])
894 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
895 """Turn the encrypted s field into a working signature"""
897 if player_url is None:
898 raise ExtractorError('Cannot decrypt signature without player_url')
900 if player_url.startswith('//'):
901 player_url = 'https:' + player_url
903 player_id = (player_url, self._signature_cache_id(s))
904 if player_id not in self._player_cache:
905 func = self._extract_signature_function(
906 video_id, player_url, s
908 self._player_cache[player_id] = func
909 func = self._player_cache[player_id]
910 if self._downloader.params.get('youtube_print_sig_code'):
911 self._print_sig_code(func, s)
913 except Exception as e:
914 tb = traceback.format_exc()
915 raise ExtractorError(
916 'Signature extraction failed: ' + tb, cause=e)
918 def _get_subtitles(self, video_id, webpage):
920 subs_doc = self._download_xml(
921 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
922 video_id, note=False)
923 except ExtractorError as err:
924 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
928 for track in subs_doc.findall('track'):
929 lang = track.attrib['lang_code']
930 if lang in sub_lang_list:
933 for ext in self._SUBTITLE_FORMATS:
934 params = compat_urllib_parse.urlencode({
938 'name': track.attrib['name'].encode('utf-8'),
941 'url': 'https://www.youtube.com/api/timedtext?' + params,
944 sub_lang_list[lang] = sub_formats
945 if not sub_lang_list:
946 self._downloader.report_warning('video doesn\'t have subtitles')
950 def _get_ytplayer_config(self, video_id, webpage):
952 # User data may contain arbitrary character sequences that may affect
953 # JSON extraction with regex, e.g. when '};' is contained the second
954 # regex won't capture the whole JSON. Yet working around by trying more
955 # concrete regex first keeping in mind proper quoted string handling
956 # to be implemented in future that will replace this workaround (see
957 # https://github.com/rg3/youtube-dl/issues/7468,
958 # https://github.com/rg3/youtube-dl/pull/7599)
959 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
960 r';ytplayer\.config\s*=\s*({.+?});',
962 config = self._search_regex(
963 patterns, webpage, 'ytplayer.config', default=None)
965 return self._parse_json(
966 uppercase_escape(config), video_id, fatal=False)
968 def _get_automatic_captions(self, video_id, webpage):
969 """We need the webpage for getting the captions url, pass it as an
970 argument to speed up the process."""
971 self.to_screen('%s: Looking for automatic captions' % video_id)
972 player_config = self._get_ytplayer_config(video_id, webpage)
973 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
974 if not player_config:
975 self._downloader.report_warning(err_msg)
978 args = player_config['args']
979 caption_url = args.get('ttsurl')
981 timestamp = args['timestamp']
982 # We get the available subtitles
983 list_params = compat_urllib_parse.urlencode({
988 list_url = caption_url + '&' + list_params
989 caption_list = self._download_xml(list_url, video_id)
990 original_lang_node = caption_list.find('track')
991 if original_lang_node is None:
992 self._downloader.report_warning('Video doesn\'t have automatic captions')
994 original_lang = original_lang_node.attrib['lang_code']
995 caption_kind = original_lang_node.attrib.get('kind', '')
998 for lang_node in caption_list.findall('target'):
999 sub_lang = lang_node.attrib['lang_code']
1001 for ext in self._SUBTITLE_FORMATS:
1002 params = compat_urllib_parse.urlencode({
1003 'lang': original_lang,
1007 'kind': caption_kind,
1009 sub_formats.append({
1010 'url': caption_url + '&' + params,
1013 sub_lang_list[sub_lang] = sub_formats
1014 return sub_lang_list
1016 # Some videos don't provide ttsurl but rather caption_tracks and
1017 # caption_translation_languages (e.g. 20LmZk1hakA)
1018 caption_tracks = args['caption_tracks']
1019 caption_translation_languages = args['caption_translation_languages']
1020 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1021 parsed_caption_url = compat_urlparse.urlparse(caption_url)
1022 caption_qs = compat_parse_qs(parsed_caption_url.query)
1025 for lang in caption_translation_languages.split(','):
1026 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1027 sub_lang = lang_qs.get('lc', [None])[0]
1031 for ext in self._SUBTITLE_FORMATS:
1033 'tlang': [sub_lang],
1036 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1037 query=compat_urllib_parse.urlencode(caption_qs, True)))
1038 sub_formats.append({
1042 sub_lang_list[sub_lang] = sub_formats
1043 return sub_lang_list
1044 # An extractor error can be raise by the download process if there are
1045 # no automatic captions but there are subtitles
1046 except (KeyError, ExtractorError):
1047 self._downloader.report_warning(err_msg)
1050 def _mark_watched(self, video_id, video_info):
1051 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1052 if not playback_url:
1054 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1055 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1057 # cpn generation algorithm is reverse engineered from base.js.
1058 # In fact it works even with dummy cpn.
1059 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1060 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1066 playback_url = compat_urlparse.urlunparse(
1067 parsed_playback_url._replace(query=compat_urllib_parse.urlencode(qs, True)))
1069 self._download_webpage(
1070 playback_url, video_id, 'Marking watched',
1071 'Unable to mark watched', fatal=False)
1074 def extract_id(cls, url):
1075 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1077 raise ExtractorError('Invalid URL: %s' % url)
1078 video_id = mobj.group(2)
1081 def _extract_from_m3u8(self, manifest_url, video_id):
1084 def _get_urls(_manifest):
1085 lines = _manifest.split('\n')
1086 urls = filter(lambda l: l and not l.startswith('#'),
1089 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1090 formats_urls = _get_urls(manifest)
1091 for format_url in formats_urls:
1092 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1093 url_map[itag] = format_url
1096 def _extract_annotations(self, video_id):
1097 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1098 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1100 def _real_extract(self, url):
1101 url, smuggled_data = unsmuggle_url(url, {})
1104 'http' if self._downloader.params.get('prefer_insecure', False)
1109 parsed_url = compat_urllib_parse_urlparse(url)
1110 for component in [parsed_url.fragment, parsed_url.query]:
1111 query = compat_parse_qs(component)
1112 if start_time is None and 't' in query:
1113 start_time = parse_duration(query['t'][0])
1114 if start_time is None and 'start' in query:
1115 start_time = parse_duration(query['start'][0])
1116 if end_time is None and 'end' in query:
1117 end_time = parse_duration(query['end'][0])
1119 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1120 mobj = re.search(self._NEXT_URL_RE, url)
1122 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1123 video_id = self.extract_id(url)
1126 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1127 video_webpage = self._download_webpage(url, video_id)
1129 # Attempt to extract SWF player URL
1130 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1131 if mobj is not None:
1132 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1138 def add_dash_mpd(video_info):
1139 dash_mpd = video_info.get('dashmpd')
1140 if dash_mpd and dash_mpd[0] not in dash_mpds:
1141 dash_mpds.append(dash_mpd[0])
1144 embed_webpage = None
1146 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1148 # We simulate the access to the video from www.youtube.com/v/{video_id}
1149 # this can be viewed without login into Youtube
1150 url = proto + '://www.youtube.com/embed/%s' % video_id
1151 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1152 data = compat_urllib_parse.urlencode({
1153 'video_id': video_id,
1154 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1155 'sts': self._search_regex(
1156 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1158 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1159 video_info_webpage = self._download_webpage(
1160 video_info_url, video_id,
1161 note='Refetching age-gated info webpage',
1162 errnote='unable to download video info webpage')
1163 video_info = compat_parse_qs(video_info_webpage)
1164 add_dash_mpd(video_info)
1168 # Try looking directly into the video webpage
1169 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1171 args = ytplayer_config['args']
1172 if args.get('url_encoded_fmt_stream_map'):
1173 # Convert to the same format returned by compat_parse_qs
1174 video_info = dict((k, [v]) for k, v in args.items())
1175 add_dash_mpd(video_info)
1176 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1178 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1179 # We also try looking in get_video_info since it may contain different dashmpd
1180 # URL that points to a DASH manifest with possibly different itag set (some itags
1181 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1182 # manifest pointed by get_video_info's dashmpd).
1183 # The general idea is to take a union of itags of both DASH manifests (for example
1184 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1185 self.report_video_info_webpage_download(video_id)
1186 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1188 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1189 % (proto, video_id, el_type))
1190 video_info_webpage = self._download_webpage(
1192 video_id, note=False,
1193 errnote='unable to download video info webpage')
1194 get_video_info = compat_parse_qs(video_info_webpage)
1195 if get_video_info.get('use_cipher_signature') != ['True']:
1196 add_dash_mpd(get_video_info)
1198 video_info = get_video_info
1199 if 'token' in get_video_info:
1200 # Different get_video_info requests may report different results, e.g.
1201 # some may report video unavailability, but some may serve it without
1202 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1203 # the original webpage as well as el=info and el=embedded get_video_info
1204 # requests report video unavailability due to geo restriction while
1205 # el=detailpage succeeds and returns valid data). This is probably
1206 # due to YouTube measures against IP ranges of hosting providers.
1207 # Working around by preferring the first succeeded video_info containing
1208 # the token if no such video_info yet was found.
1209 if 'token' not in video_info:
1210 video_info = get_video_info
1212 if 'token' not in video_info:
1213 if 'reason' in video_info:
1214 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1215 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1217 raise ExtractorError('YouTube said: This video is available in %s only' % (
1218 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1220 raise ExtractorError(
1221 'YouTube said: %s' % video_info['reason'][0],
1222 expected=True, video_id=video_id)
1224 raise ExtractorError(
1225 '"token" parameter not in video info for unknown reason',
1229 if 'title' in video_info:
1230 video_title = video_info['title'][0]
1232 self._downloader.report_warning('Unable to extract video title')
1236 video_description = get_element_by_id("eow-description", video_webpage)
1237 if video_description:
1238 video_description = re.sub(r'''(?x)
1240 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1241 (?:title|href)="([^"]+)"\s+
1242 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1243 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1246 ''', r'\1', video_description)
1247 video_description = clean_html(video_description)
1249 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1251 video_description = unescapeHTML(fd_mobj.group(1))
1253 video_description = ''
1255 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1256 if not self._downloader.params.get('noplaylist'):
1259 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1260 for feed in multifeed_metadata_list.split(','):
1261 # Unquote should take place before split on comma (,) since textual
1262 # fields may contain comma as well (see
1263 # https://github.com/rg3/youtube-dl/issues/8536)
1264 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1266 '_type': 'url_transparent',
1267 'ie_key': 'Youtube',
1269 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1270 {'force_singlefeed': True}),
1271 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1273 feed_ids.append(feed_data['id'][0])
1275 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1276 % (', '.join(feed_ids), video_id))
1277 return self.playlist_result(entries, video_id, video_title, video_description)
1278 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1280 if 'view_count' in video_info:
1281 view_count = int(video_info['view_count'][0])
1285 # Check for "rental" videos
1286 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1287 raise ExtractorError('"rental" videos not supported')
1289 # Start extracting information
1290 self.report_information_extraction(video_id)
1293 if 'author' not in video_info:
1294 raise ExtractorError('Unable to extract uploader name')
1295 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1298 video_uploader_id = None
1299 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1300 if mobj is not None:
1301 video_uploader_id = mobj.group(1)
1303 self._downloader.report_warning('unable to extract uploader nickname')
1306 # We try first to get a high quality image:
1307 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1308 video_webpage, re.DOTALL)
1309 if m_thumb is not None:
1310 video_thumbnail = m_thumb.group(1)
1311 elif 'thumbnail_url' not in video_info:
1312 self._downloader.report_warning('unable to extract video thumbnail')
1313 video_thumbnail = None
1314 else: # don't panic if we can't find it
1315 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1318 upload_date = self._html_search_meta(
1319 'datePublished', video_webpage, 'upload date', default=None)
1321 upload_date = self._search_regex(
1322 [r'(?s)id="eow-date.*?>(.*?)</span>',
1323 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1324 video_webpage, 'upload date', default=None)
1326 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1327 upload_date = unified_strdate(upload_date)
1329 m_music = re.search(
1330 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1333 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1334 video_creator = clean_html(m_music.group('creator'))
1336 video_alt_title = video_creator = None
1338 m_cat_container = self._search_regex(
1339 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1340 video_webpage, 'categories', default=None)
1342 category = self._html_search_regex(
1343 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1345 video_categories = None if category is None else [category]
1347 video_categories = None
1350 unescapeHTML(m.group('content'))
1351 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1353 def _extract_count(count_name):
1354 return str_to_int(self._search_regex(
1355 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1356 % re.escape(count_name),
1357 video_webpage, count_name, default=None))
1359 like_count = _extract_count('like')
1360 dislike_count = _extract_count('dislike')
1363 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1364 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1366 if 'length_seconds' not in video_info:
1367 self._downloader.report_warning('unable to extract video duration')
1368 video_duration = None
1370 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1373 video_annotations = None
1374 if self._downloader.params.get('writeannotations', False):
1375 video_annotations = self._extract_annotations(video_id)
1377 def _map_to_format_list(urlmap):
1379 for itag, video_real_url in urlmap.items():
1382 'url': video_real_url,
1383 'player_url': player_url,
1385 if itag in self._formats:
1386 dct.update(self._formats[itag])
1390 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1391 self.report_rtmp_download()
1393 'format_id': '_rtmp',
1395 'url': video_info['conn'][0],
1396 'player_url': player_url,
1398 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1399 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1400 if 'rtmpe%3Dyes' in encoded_url_map:
1401 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1403 fmt_list = video_info.get('fmt_list', [''])[0]
1405 for fmt in fmt_list.split(','):
1406 spec = fmt.split('/')
1408 width_height = spec[1].split('x')
1409 if len(width_height) == 2:
1410 formats_spec[spec[0]] = {
1411 'resolution': spec[1],
1412 'width': int_or_none(width_height[0]),
1413 'height': int_or_none(width_height[1]),
1416 for url_data_str in encoded_url_map.split(','):
1417 url_data = compat_parse_qs(url_data_str)
1418 if 'itag' not in url_data or 'url' not in url_data:
1420 format_id = url_data['itag'][0]
1421 url = url_data['url'][0]
1423 if 'sig' in url_data:
1424 url += '&signature=' + url_data['sig'][0]
1425 elif 's' in url_data:
1426 encrypted_sig = url_data['s'][0]
1427 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1429 jsplayer_url_json = self._search_regex(
1431 embed_webpage if age_gate else video_webpage,
1432 'JS player URL (1)', default=None)
1433 if not jsplayer_url_json and not age_gate:
1434 # We need the embed website after all
1435 if embed_webpage is None:
1436 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1437 embed_webpage = self._download_webpage(
1438 embed_url, video_id, 'Downloading embed webpage')
1439 jsplayer_url_json = self._search_regex(
1440 ASSETS_RE, embed_webpage, 'JS player URL')
1442 player_url = json.loads(jsplayer_url_json)
1443 if player_url is None:
1444 player_url_json = self._search_regex(
1445 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1446 video_webpage, 'age gate player URL')
1447 player_url = json.loads(player_url_json)
1449 if self._downloader.params.get('verbose'):
1450 if player_url is None:
1451 player_version = 'unknown'
1452 player_desc = 'unknown'
1454 if player_url.endswith('swf'):
1455 player_version = self._search_regex(
1456 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1457 'flash player', fatal=False)
1458 player_desc = 'flash player %s' % player_version
1460 player_version = self._search_regex(
1461 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1463 'html5 player', fatal=False)
1464 player_desc = 'html5 player %s' % player_version
1466 parts_sizes = self._signature_cache_id(encrypted_sig)
1467 self.to_screen('{%s} signature length %s, %s' %
1468 (format_id, parts_sizes, player_desc))
1470 signature = self._decrypt_signature(
1471 encrypted_sig, video_id, player_url, age_gate)
1472 url += '&signature=' + signature
1473 if 'ratebypass' not in url:
1474 url += '&ratebypass=yes'
1477 'format_id': format_id,
1479 'player_url': player_url,
1481 if format_id in self._formats:
1482 dct.update(self._formats[format_id])
1483 if format_id in formats_spec:
1484 dct.update(formats_spec[format_id])
1486 # Some itags are not included in DASH manifest thus corresponding formats will
1487 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1488 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1489 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1490 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1493 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1494 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1497 'fps': int_or_none(url_data.get('fps', [None])[0]),
1498 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1500 for key, value in more_fields.items():
1503 type_ = url_data.get('type', [None])[0]
1505 type_split = type_.split(';')
1506 kind_ext = type_split[0].split('/')
1507 if len(kind_ext) == 2:
1509 dct['ext'] = mimetype2ext(type_split[0])
1510 if kind in ('audio', 'video'):
1512 for mobj in re.finditer(
1513 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1514 if mobj.group('key') == 'codecs':
1515 codecs = mobj.group('val')
1518 codecs = codecs.split(',')
1519 if len(codecs) == 2:
1520 acodec, vcodec = codecs[1], codecs[0]
1522 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1528 elif video_info.get('hlsvp'):
1529 manifest_url = video_info['hlsvp'][0]
1530 url_map = self._extract_from_m3u8(manifest_url, video_id)
1531 formats = _map_to_format_list(url_map)
1532 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1533 for a_format in formats:
1534 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1536 unavailable_message = self._html_search_regex(
1537 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1538 video_webpage, 'unavailable message', default=None)
1539 if unavailable_message:
1540 raise ExtractorError(unavailable_message, expected=True)
1541 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1543 # Look for the DASH manifest
1544 if self._downloader.params.get('youtube_include_dash_manifest', True):
1545 dash_mpd_fatal = True
1546 for mpd_url in dash_mpds:
1549 def decrypt_sig(mobj):
1551 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1552 return '/signature/%s' % dec_s
1554 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1556 for df in self._extract_mpd_formats(
1557 mpd_url, video_id, fatal=dash_mpd_fatal,
1558 formats_dict=self._formats):
1559 # Do not overwrite DASH format found in some previous DASH manifest
1560 if df['format_id'] not in dash_formats:
1561 dash_formats[df['format_id']] = df
1562 # Additional DASH manifests may end up in HTTP Error 403 therefore
1563 # allow them to fail without bug report message if we already have
1564 # some DASH manifest succeeded. This is temporary workaround to reduce
1565 # burst of bug reports until we figure out the reason and whether it
1566 # can be fixed at all.
1567 dash_mpd_fatal = False
1568 except (ExtractorError, KeyError) as e:
1569 self.report_warning(
1570 'Skipping DASH manifest: %r' % e, video_id)
1572 # Remove the formats we found through non-DASH, they
1573 # contain less info and it can be wrong, because we use
1574 # fixed values (for example the resolution). See
1575 # https://github.com/rg3/youtube-dl/issues/5774 for an
1577 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1578 formats.extend(dash_formats.values())
1580 # Check for malformed aspect ratio
1581 stretched_m = re.search(
1582 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1585 w = float(stretched_m.group('w'))
1586 h = float(stretched_m.group('h'))
1587 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1588 # We will only process correct ratios.
1592 if f.get('vcodec') != 'none':
1593 f['stretched_ratio'] = ratio
1595 self._sort_formats(formats)
1597 self.mark_watched(video_id, video_info)
1601 'uploader': video_uploader,
1602 'uploader_id': video_uploader_id,
1603 'upload_date': upload_date,
1604 'creator': video_creator,
1605 'title': video_title,
1606 'alt_title': video_alt_title,
1607 'thumbnail': video_thumbnail,
1608 'description': video_description,
1609 'categories': video_categories,
1611 'subtitles': video_subtitles,
1612 'automatic_captions': automatic_captions,
1613 'duration': video_duration,
1614 'age_limit': 18 if age_gate else 0,
1615 'annotations': video_annotations,
1616 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1617 'view_count': view_count,
1618 'like_count': like_count,
1619 'dislike_count': dislike_count,
1620 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1623 'start_time': start_time,
1624 'end_time': end_time,
1628 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1629 IE_DESC = 'YouTube.com playlists'
1630 _VALID_URL = r"""(?x)(?:
1635 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1636 \? (?:.*?[&;])*? (?:p|a|list)=
1640 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1641 # Top tracks, they can also include dots
1646 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1648 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1649 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1650 IE_NAME = 'youtube:playlist'
1652 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1654 'title': 'ytdl test PL',
1655 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1657 'playlist_count': 3,
1659 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1661 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1662 'title': 'YDL_Empty_List',
1664 'playlist_count': 0,
1666 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1667 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1669 'title': '29C3: Not my department',
1670 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1672 'playlist_count': 95,
1674 'note': 'issue #673',
1675 'url': 'PLBB231211A4F62143',
1677 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1678 'id': 'PLBB231211A4F62143',
1680 'playlist_mincount': 26,
1682 'note': 'Large playlist',
1683 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1685 'title': 'Uploads from Cauchemar',
1686 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1688 'playlist_mincount': 799,
1690 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1692 'title': 'YDL_safe_search',
1693 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1695 'playlist_count': 2,
1698 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1699 'playlist_count': 4,
1702 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1705 'note': 'Embedded SWF player',
1706 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1707 'playlist_count': 4,
1710 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1713 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1714 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1716 'title': 'Uploads from Interstellar Movie',
1717 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1719 'playlist_mincout': 21,
1722 def _real_initialize(self):
1725 def _extract_mix(self, playlist_id):
1726 # The mixes are generated from a single video
1727 # the id of the playlist is just 'RD' + video_id
1728 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1729 webpage = self._download_webpage(
1730 url, playlist_id, 'Downloading Youtube mix')
1731 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1733 search_title('playlist-title') or
1734 search_title('title long-title') or
1735 search_title('title'))
1736 title = clean_html(title_span)
1737 ids = orderedSet(re.findall(
1738 r'''(?xs)data-video-username=".*?".*?
1739 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1741 url_results = self._ids_to_results(ids)
1743 return self.playlist_result(url_results, playlist_id, title)
1745 def _extract_playlist(self, playlist_id):
1746 url = self._TEMPLATE_URL % playlist_id
1747 page = self._download_webpage(url, playlist_id)
1749 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1750 match = match.strip()
1751 # Check if the playlist exists or is private
1752 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1753 raise ExtractorError(
1754 'The playlist doesn\'t exist or is private, use --username or '
1755 '--netrc to access it.',
1757 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1758 raise ExtractorError(
1759 'Invalid parameters. Maybe URL is incorrect.',
1761 elif re.match(r'[^<]*Choose your language[^<]*', match):
1764 self.report_warning('Youtube gives an alert message: ' + match)
1766 playlist_title = self._html_search_regex(
1767 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1770 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1772 def _check_download_just_video(self, url, playlist_id):
1773 # Check if it's a video-specific URL
1774 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1775 if 'v' in query_dict:
1776 video_id = query_dict['v'][0]
1777 if self._downloader.params.get('noplaylist'):
1778 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1779 return self.url_result(video_id, 'Youtube', video_id=video_id)
1781 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1783 def _real_extract(self, url):
1784 # Extract playlist id
1785 mobj = re.match(self._VALID_URL, url)
1787 raise ExtractorError('Invalid URL: %s' % url)
1788 playlist_id = mobj.group(1) or mobj.group(2)
1790 video = self._check_download_just_video(url, playlist_id)
1794 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1795 # Mixes require a custom extraction process
1796 return self._extract_mix(playlist_id)
1798 return self._extract_playlist(playlist_id)
1801 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1802 IE_DESC = 'YouTube.com channels'
1803 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1804 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1805 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1806 IE_NAME = 'youtube:channel'
1808 'note': 'paginated channel',
1809 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1810 'playlist_mincount': 91,
1812 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1813 'title': 'Uploads from lex will',
1816 'note': 'Age restricted channel',
1817 # from https://www.youtube.com/user/DeusExOfficial
1818 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1819 'playlist_mincount': 64,
1821 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1822 'title': 'Uploads from Deus Ex',
1827 def suitable(cls, url):
1828 return False if YoutubePlaylistsIE.suitable(url) else super(YoutubeChannelIE, cls).suitable(url)
1830 def _real_extract(self, url):
1831 channel_id = self._match_id(url)
1833 url = self._TEMPLATE_URL % channel_id
1835 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1836 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1837 # otherwise fallback on channel by page extraction
1838 channel_page = self._download_webpage(
1839 url + '?view=57', channel_id,
1840 'Downloading channel page', fatal=False)
1841 if channel_page is False:
1842 channel_playlist_id = False
1844 channel_playlist_id = self._html_search_meta(
1845 'channelId', channel_page, 'channel id', default=None)
1846 if not channel_playlist_id:
1847 channel_playlist_id = self._search_regex(
1848 r'data-(?:channel-external-|yt)id="([^"]+)"',
1849 channel_page, 'channel id', default=None)
1850 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1851 playlist_id = 'UU' + channel_playlist_id[2:]
1852 return self.url_result(
1853 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1855 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1856 autogenerated = re.search(r'''(?x)
1858 channel-header-autogenerated-label|
1859 yt-channel-title-autogenerated
1860 )[^"]*"''', channel_page) is not None
1863 # The videos are contained in a single page
1864 # the ajax pages can't be used, they are empty
1867 video_id, 'Youtube', video_id=video_id,
1868 video_title=video_title)
1869 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1870 return self.playlist_result(entries, channel_id)
1872 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1875 class YoutubeUserIE(YoutubeChannelIE):
1876 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1877 _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_-]+)'
1878 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1879 IE_NAME = 'youtube:user'
1882 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1883 'playlist_mincount': 320,
1885 'title': 'TheLinuxFoundation',
1888 'url': 'ytuser:phihag',
1889 'only_matching': True,
1893 def suitable(cls, url):
1894 # Don't return True if the url can be extracted with other youtube
1895 # extractor, the regex would is too permissive and it would match.
1896 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1897 if any(ie.suitable(url) for ie in other_ies):
1900 return super(YoutubeUserIE, cls).suitable(url)
1903 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1904 IE_DESC = 'YouTube.com user/channel playlists'
1905 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
1906 IE_NAME = 'youtube:playlists'
1909 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1910 'playlist_mincount': 4,
1912 'id': 'ThirstForScience',
1913 'title': 'Thirst for Science',
1916 # with "Load more" button
1917 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1918 'playlist_mincount': 70,
1921 'title': 'Игорь Клейнер',
1924 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
1925 'playlist_mincount': 17,
1927 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
1928 'title': 'Chem Player',
1933 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1934 IE_DESC = 'YouTube.com searches'
1935 # there doesn't appear to be a real limit, for example if you search for
1936 # 'python' you get more than 8.000.000 results
1937 _MAX_RESULTS = float('inf')
1938 IE_NAME = 'youtube:search'
1939 _SEARCH_KEY = 'ytsearch'
1940 _EXTRA_QUERY_ARGS = {}
1943 def _get_n_results(self, query, n):
1944 """Get a specified number of results for a query"""
1949 for pagenum in itertools.count(1):
1951 'search_query': query.encode('utf-8'),
1955 url_query.update(self._EXTRA_QUERY_ARGS)
1956 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1957 data = self._download_json(
1958 result_url, video_id='query "%s"' % query,
1959 note='Downloading page %s' % pagenum,
1960 errnote='Unable to download API page')
1961 html_content = data[1]['body']['content']
1963 if 'class="search-message' in html_content:
1964 raise ExtractorError(
1965 '[youtube] No video results', expected=True)
1967 new_videos = self._ids_to_results(orderedSet(re.findall(
1968 r'href="/watch\?v=(.{11})', html_content)))
1969 videos += new_videos
1970 if not new_videos or len(videos) > limit:
1975 return self.playlist_result(videos, query)
1978 class YoutubeSearchDateIE(YoutubeSearchIE):
1979 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1980 _SEARCH_KEY = 'ytsearchdate'
1981 IE_DESC = 'YouTube.com searches, newest videos first'
1982 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1985 class YoutubeSearchURLIE(InfoExtractor):
1986 IE_DESC = 'YouTube.com search URLs'
1987 IE_NAME = 'youtube:search_url'
1988 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
1990 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1991 'playlist_mincount': 5,
1993 'title': 'youtube-dl test video',
1996 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
1997 'only_matching': True,
2000 def _real_extract(self, url):
2001 mobj = re.match(self._VALID_URL, url)
2002 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2004 webpage = self._download_webpage(url, query)
2005 result_code = self._search_regex(
2006 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
2008 part_codes = re.findall(
2009 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
2011 for part_code in part_codes:
2012 part_title = self._html_search_regex(
2013 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
2014 part_url_snippet = self._html_search_regex(
2015 r'(?s)href="([^"]+)"', part_code, 'item URL')
2016 part_url = compat_urlparse.urljoin(
2017 'https://www.youtube.com/', part_url_snippet)
2021 'title': part_title,
2025 '_type': 'playlist',
2031 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2032 IE_DESC = 'YouTube.com (multi-season) shows'
2033 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2034 IE_NAME = 'youtube:show'
2036 'url': 'https://www.youtube.com/show/airdisasters',
2037 'playlist_mincount': 5,
2039 'id': 'airdisasters',
2040 'title': 'Air Disasters',
2044 def _real_extract(self, url):
2045 playlist_id = self._match_id(url)
2046 return super(YoutubeShowIE, self)._real_extract(
2047 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2050 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2052 Base class for feed extractors
2053 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2055 _LOGIN_REQUIRED = True
2059 return 'youtube:%s' % self._FEED_NAME
2061 def _real_initialize(self):
2064 def _real_extract(self, url):
2065 page = self._download_webpage(
2066 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2068 # The extraction process is the same as for playlists, but the regex
2069 # for the video ids doesn't contain an index
2071 more_widget_html = content_html = page
2072 for page_num in itertools.count(1):
2073 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2075 # 'recommended' feed has infinite 'load more' and each new portion spins
2076 # the same videos in (sometimes) slightly different order, so we'll check
2077 # for unicity and break when portion has no new videos
2078 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2084 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2088 more = self._download_json(
2089 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2090 'Downloading page #%s' % page_num,
2091 transform_source=uppercase_escape)
2092 content_html = more['content_html']
2093 more_widget_html = more['load_more_widget_html']
2095 return self.playlist_result(
2096 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2099 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2100 IE_NAME = 'youtube:watchlater'
2101 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2102 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2105 'url': 'https://www.youtube.com/playlist?list=WL',
2106 'only_matching': True,
2108 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2109 'only_matching': True,
2112 def _real_extract(self, url):
2113 video = self._check_download_just_video(url, 'WL')
2116 return self._extract_playlist('WL')
2119 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2120 IE_NAME = 'youtube:favorites'
2121 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2122 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2123 _LOGIN_REQUIRED = True
2125 def _real_extract(self, url):
2126 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2127 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2128 return self.url_result(playlist_id, 'YoutubePlaylist')
2131 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2132 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2133 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2134 _FEED_NAME = 'recommended'
2135 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2138 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2139 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2140 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2141 _FEED_NAME = 'subscriptions'
2142 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2145 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2146 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2147 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2148 _FEED_NAME = 'history'
2149 _PLAYLIST_TITLE = 'Youtube History'
2152 class YoutubeTruncatedURLIE(InfoExtractor):
2153 IE_NAME = 'youtube:truncated_url'
2154 IE_DESC = False # Do not list
2155 _VALID_URL = r'''(?x)
2157 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2160 annotation_id=annotation_[^&]+|
2166 attribution_link\?a=[^&]+
2172 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2173 'only_matching': True,
2175 'url': 'http://www.youtube.com/watch?',
2176 'only_matching': True,
2178 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2179 'only_matching': True,
2181 'url': 'https://www.youtube.com/watch?feature=foo',
2182 'only_matching': True,
2184 'url': 'https://www.youtube.com/watch?hl=en-GB',
2185 'only_matching': True,
2187 'url': 'https://www.youtube.com/watch?t=2372',
2188 'only_matching': True,
2191 def _real_extract(self, url):
2192 raise ExtractorError(
2193 'Did you forget to quote the URL? Remember that & is a meta '
2194 'character in most shells, so you want to put the URL in quotes, '
2196 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2197 ' or simply youtube-dl BaW_jenozKc .',
2201 class YoutubeTruncatedIDIE(InfoExtractor):
2202 IE_NAME = 'youtube:truncated_id'
2203 IE_DESC = False # Do not list
2204 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2207 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2208 'only_matching': True,
2211 def _real_extract(self, url):
2212 video_id = self._match_id(url)
2213 raise ExtractorError(
2214 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),