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,
31 get_element_by_attribute,
48 class YoutubeBaseInfoExtractor(InfoExtractor):
49 """Provide base functions for Youtube extractors"""
50 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
51 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
52 _NETRC_MACHINE = 'youtube'
53 # If True it will raise an error if no login info is provided
54 _LOGIN_REQUIRED = False
56 def _set_language(self):
58 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
59 # YouTube sets the expire time to about two months
60 expire_time=time.time() + 2 * 30 * 24 * 3600)
62 def _ids_to_results(self, ids):
64 self.url_result(vid_id, 'Youtube', video_id=vid_id)
69 Attempt to log in to YouTube.
70 True is returned if successful or skipped.
71 False is returned if login failed.
73 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
75 (username, password) = self._get_login_info()
76 # No authentication to be performed
78 if self._LOGIN_REQUIRED:
79 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
82 login_page = self._download_webpage(
83 self._LOGIN_URL, None,
84 note='Downloading login page',
85 errnote='unable to fetch login page', fatal=False)
86 if login_page is False:
89 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
90 login_page, 'Login GALX parameter')
94 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
99 'PersistentCookie': 'yes',
101 'bgresponse': 'js_disabled',
102 'checkConnection': '',
103 'checkedDomains': 'youtube',
110 'service': 'youtube',
115 login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii')
117 req = sanitized_Request(self._LOGIN_URL, login_data)
118 login_results = self._download_webpage(
120 note='Logging in', errnote='unable to log in', fatal=False)
121 if login_results is False:
124 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
125 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
128 # TODO add SMS and phone call support - these require making a request and then prompting the user
130 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
131 tfa_code = self._get_tfa_info('2-step verification code')
134 self._downloader.report_warning(
135 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
136 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
139 tfa_code = remove_start(tfa_code, 'G-')
141 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
143 tfa_form_strs.update({
148 tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii')
150 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
151 tfa_results = self._download_webpage(
153 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
155 if tfa_results is False:
158 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
159 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
161 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
162 self._downloader.report_warning('unable to log in - did the page structure change?')
164 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
165 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
168 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
169 self._downloader.report_warning('unable to log in: bad username or password')
173 def _real_initialize(self):
174 if self._downloader is None:
177 if not self._login():
181 class YoutubeEntryListBaseInfoExtractor(InfoExtractor):
182 # Extract entries from page with "Load more" button
183 def _entries(self, page, playlist_id):
184 more_widget_html = content_html = page
185 for page_num in itertools.count(1):
186 for entry in self._process_page(content_html):
189 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
193 more = self._download_json(
194 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
195 'Downloading page #%s' % page_num,
196 transform_source=uppercase_escape)
197 content_html = more['content_html']
198 if not content_html.strip():
199 # Some webpages show a "Load more" button but they don't
202 more_widget_html = more['load_more_widget_html']
205 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
206 def _process_page(self, content):
207 for video_id, video_title in self.extract_videos_from_page(content):
208 yield self.url_result(video_id, 'Youtube', video_id, video_title)
210 def extract_videos_from_page(self, page):
213 for mobj in re.finditer(self._VIDEO_RE, page):
214 # The link with index 0 is not the first video of the playlist (not sure if still actual)
215 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
217 video_id = mobj.group('id')
218 video_title = unescapeHTML(mobj.group('title'))
220 video_title = video_title.strip()
222 idx = ids_in_page.index(video_id)
223 if video_title and not titles_in_page[idx]:
224 titles_in_page[idx] = video_title
226 ids_in_page.append(video_id)
227 titles_in_page.append(video_title)
228 return zip(ids_in_page, titles_in_page)
231 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
232 def _process_page(self, content):
233 for playlist_id in re.findall(r'href="/?playlist\?list=(.+?)"', content):
234 yield self.url_result(
235 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
237 def _real_extract(self, url):
238 playlist_id = self._match_id(url)
239 webpage = self._download_webpage(url, playlist_id)
240 title = self._og_search_title(webpage, fatal=False)
241 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
244 class YoutubeIE(YoutubeBaseInfoExtractor):
245 IE_DESC = 'YouTube.com'
246 _VALID_URL = r"""(?x)^
248 (?:https?://|//) # http(s):// or protocol-independent URL
249 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
250 (?:www\.)?deturl\.com/www\.youtube\.com/|
251 (?:www\.)?pwnyoutube\.com/|
252 (?:www\.)?yourepeat\.com/|
253 tube\.majestyc\.net/|
254 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
255 (?:.*?\#/)? # handle anchor (#/) redirect urls
256 (?: # the various things that can precede the ID:
257 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
258 |(?: # or the v= param in all its forms
259 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
260 (?:\?|\#!?) # the params delimiter ? or # or #!
261 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
266 youtu\.be| # just youtu.be/xxxx
267 vid\.plus # or vid.plus/xxxx
269 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
271 )? # all until now is optional -> you can pass the naked ID
272 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
273 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
274 (?(1).+)? # if we found the ID, everything can follow
276 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
278 '5': {'ext': 'flv', 'width': 400, 'height': 240},
279 '6': {'ext': 'flv', 'width': 450, 'height': 270},
280 '13': {'ext': '3gp'},
281 '17': {'ext': '3gp', 'width': 176, 'height': 144},
282 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
283 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
284 '34': {'ext': 'flv', 'width': 640, 'height': 360},
285 '35': {'ext': 'flv', 'width': 854, 'height': 480},
286 '36': {'ext': '3gp', 'width': 320, 'height': 240},
287 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
288 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
289 '43': {'ext': 'webm', 'width': 640, 'height': 360},
290 '44': {'ext': 'webm', 'width': 854, 'height': 480},
291 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
292 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
293 '59': {'ext': 'mp4', 'width': 854, 'height': 480},
294 '78': {'ext': 'mp4', 'width': 854, 'height': 480},
298 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
299 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
300 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
301 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
302 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
303 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
304 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
306 # Apple HTTP Live Streaming
307 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
308 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
309 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
310 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
311 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
312 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
313 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
316 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
317 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
318 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
319 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
320 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
321 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
322 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
323 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
324 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
325 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
326 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
329 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
330 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
331 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
334 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
335 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
336 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
337 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
338 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
339 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
340 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'vp9'},
341 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
342 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
343 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
344 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
345 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
346 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
347 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
348 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
349 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
350 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
351 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
352 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
353 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
354 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'vp9'},
355 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
358 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
359 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
361 # Dash webm audio with opus inside
362 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
363 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
364 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
367 '_rtmp': {'protocol': 'rtmp'},
373 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
377 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
378 'uploader': 'Philipp Hagemeister',
379 'uploader_id': 'phihag',
380 'upload_date': '20121002',
381 '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 .',
382 'categories': ['Science & Technology'],
383 'tags': ['youtube-dl'],
385 'dislike_count': int,
391 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
392 'note': 'Test generic use_cipher_signature video (#897)',
396 'upload_date': '20120506',
397 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
398 'description': 'md5:782e8651347686cba06e58f71ab51773',
399 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
400 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
401 'iconic ep', 'iconic', 'love', 'it'],
402 'uploader': 'Icona Pop',
403 'uploader_id': 'IconaPop',
407 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
408 'note': 'Test VEVO video with age protection (#956)',
412 'upload_date': '20130703',
413 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
414 'description': 'md5:64249768eec3bc4276236606ea996373',
415 'uploader': 'justintimberlakeVEVO',
416 'uploader_id': 'justintimberlakeVEVO',
421 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
422 'note': 'Embed-only video (#1746)',
426 'upload_date': '20120608',
427 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
428 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
429 'uploader': 'SET India',
430 'uploader_id': 'setindia',
435 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY',
436 'note': 'Use the first video ID in the URL',
440 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
441 'uploader': 'Philipp Hagemeister',
442 'uploader_id': 'phihag',
443 'upload_date': '20121002',
444 '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 .',
445 'categories': ['Science & Technology'],
446 'tags': ['youtube-dl'],
448 'dislike_count': int,
451 'skip_download': True,
455 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
456 'note': '256k DASH audio (format 141) via DASH manifest',
460 'upload_date': '20121002',
461 'uploader_id': '8KVIDEO',
463 'uploader': '8KVIDEO',
464 'title': 'UHDTV TEST 8K VIDEO.mp4'
467 'youtube_include_dash_manifest': True,
471 # DASH manifest with encrypted signature
473 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
477 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
478 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
479 'uploader': 'AfrojackVEVO',
480 'uploader_id': 'AfrojackVEVO',
481 'upload_date': '20131011',
484 'youtube_include_dash_manifest': True,
488 # JS player signature function name containing $
490 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
494 'title': 'Taylor Swift - Shake It Off',
495 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
496 'uploader': 'TaylorSwiftVEVO',
497 'uploader_id': 'TaylorSwiftVEVO',
498 'upload_date': '20140818',
501 'youtube_include_dash_manifest': True,
507 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
511 'upload_date': '20100909',
512 'uploader': 'The Amazing Atheist',
513 'uploader_id': 'TheAmazingAtheist',
514 'title': 'Burning Everyone\'s Koran',
515 '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',
518 # Normal age-gate video (No vevo, embed allowed)
520 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
524 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
525 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
526 'uploader': 'The Witcher',
527 'uploader_id': 'WitcherGame',
528 'upload_date': '20140605',
532 # Age-gate video with encrypted signature
534 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
538 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
539 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
540 'uploader': 'LloydVEVO',
541 'uploader_id': 'LloydVEVO',
542 'upload_date': '20110629',
546 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
548 'url': '__2ABJjxzNo',
552 'upload_date': '20100430',
553 'uploader_id': 'deadmau5',
554 'description': 'md5:12c56784b8032162bb936a5f76d55360',
555 'uploader': 'deadmau5',
556 'title': 'Deadmau5 - Some Chords (HD)',
558 'expected_warnings': [
559 'DASH manifest missing',
562 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
564 'url': 'lqQg6PlCWgI',
568 'upload_date': '20150827',
569 'uploader_id': 'olympic',
570 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
571 'uploader': 'Olympics',
572 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
575 'skip_download': 'requires avconv',
580 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
584 'stretched_ratio': 16 / 9.,
585 'upload_date': '20110310',
586 'uploader_id': 'AllenMeow',
587 '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',
589 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
592 # url_encoded_fmt_stream_map is empty string
594 'url': 'qEJwOuvDf7I',
598 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
600 'upload_date': '20150404',
601 'uploader_id': 'spbelect',
602 'uploader': 'Наблюдатели Петербурга',
605 'skip_download': 'requires avconv',
608 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
610 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
614 'title': 'md5:7b81415841e02ecd4313668cde88737a',
615 'description': 'md5:116377fd2963b81ec4ce64b542173306',
616 'upload_date': '20150625',
617 'uploader_id': 'dorappi2000',
618 'uploader': 'dorappi2000',
619 'formats': 'mincount:33',
622 # DASH manifest with segment_list
624 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
625 'md5': '8ce563a1d667b599d21064e982ab9e31',
629 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
630 'uploader': 'Airtek',
631 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
632 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
633 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
636 'youtube_include_dash_manifest': True,
637 'format': '135', # bestvideo
641 # Multifeed videos (multiple cameras), URL is for Main Camera
642 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
645 'title': 'teamPGP: Rocket League Noob Stream',
646 'description': 'md5:dc7872fb300e143831327f1bae3af010',
652 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
653 'description': 'md5:dc7872fb300e143831327f1bae3af010',
654 'upload_date': '20150721',
655 'uploader': 'Beer Games Beer',
656 'uploader_id': 'beergamesbeer',
662 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
663 'description': 'md5:dc7872fb300e143831327f1bae3af010',
664 'upload_date': '20150721',
665 'uploader': 'Beer Games Beer',
666 'uploader_id': 'beergamesbeer',
672 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
673 'description': 'md5:dc7872fb300e143831327f1bae3af010',
674 'upload_date': '20150721',
675 'uploader': 'Beer Games Beer',
676 'uploader_id': 'beergamesbeer',
682 'title': 'teamPGP: Rocket League Noob Stream (zim)',
683 'description': 'md5:dc7872fb300e143831327f1bae3af010',
684 'upload_date': '20150721',
685 'uploader': 'Beer Games Beer',
686 'uploader_id': 'beergamesbeer',
690 'skip_download': True,
694 'url': 'http://vid.plus/FlRa-iH7PGw',
695 'only_matching': True,
698 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
699 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
703 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
704 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
705 'upload_date': '20151119',
706 'uploader_id': 'IronSoulElf',
707 'uploader': 'IronSoulElf',
710 'skip_download': True,
714 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
715 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
716 'only_matching': True,
719 # Video with yt:stretch=17:0
720 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
724 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
725 'description': 'md5:ee18a25c350637c8faff806845bddee9',
726 'upload_date': '20151107',
727 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
728 'uploader': 'CH GAMER DROID',
731 'skip_download': True,
735 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
736 'only_matching': True,
740 def __init__(self, *args, **kwargs):
741 super(YoutubeIE, self).__init__(*args, **kwargs)
742 self._player_cache = {}
744 def report_video_info_webpage_download(self, video_id):
745 """Report attempt to download video info webpage."""
746 self.to_screen('%s: Downloading video info webpage' % video_id)
748 def report_information_extraction(self, video_id):
749 """Report attempt to extract video information."""
750 self.to_screen('%s: Extracting video information' % video_id)
752 def report_unavailable_format(self, video_id, format):
753 """Report extracted video URL."""
754 self.to_screen('%s: Format %s not available' % (video_id, format))
756 def report_rtmp_download(self):
757 """Indicate the download will use the RTMP protocol."""
758 self.to_screen('RTMP download detected')
760 def _signature_cache_id(self, example_sig):
761 """ Return a string representation of a signature """
762 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
764 def _extract_signature_function(self, video_id, player_url, example_sig):
766 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
769 raise ExtractorError('Cannot identify player %r' % player_url)
770 player_type = id_m.group('ext')
771 player_id = id_m.group('id')
773 # Read from filesystem cache
774 func_id = '%s_%s_%s' % (
775 player_type, player_id, self._signature_cache_id(example_sig))
776 assert os.path.basename(func_id) == func_id
778 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
779 if cache_spec is not None:
780 return lambda s: ''.join(s[i] for i in cache_spec)
783 'Downloading player %s' % player_url
784 if self._downloader.params.get('verbose') else
785 'Downloading %s player %s' % (player_type, player_id)
787 if player_type == 'js':
788 code = self._download_webpage(
789 player_url, video_id,
791 errnote='Download of %s failed' % player_url)
792 res = self._parse_sig_js(code)
793 elif player_type == 'swf':
794 urlh = self._request_webpage(
795 player_url, video_id,
797 errnote='Download of %s failed' % player_url)
799 res = self._parse_sig_swf(code)
801 assert False, 'Invalid player type %r' % player_type
803 test_string = ''.join(map(compat_chr, range(len(example_sig))))
804 cache_res = res(test_string)
805 cache_spec = [ord(c) for c in cache_res]
807 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
810 def _print_sig_code(self, func, example_sig):
811 def gen_sig_code(idxs):
812 def _genslice(start, end, step):
813 starts = '' if start == 0 else str(start)
814 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
815 steps = '' if step == 1 else (':%d' % step)
816 return 's[%s%s%s]' % (starts, ends, steps)
819 # Quelch pyflakes warnings - start will be set when step is set
820 start = '(Never used)'
821 for i, prev in zip(idxs[1:], idxs[:-1]):
825 yield _genslice(start, prev, step)
828 if i - prev in [-1, 1]:
837 yield _genslice(start, i, step)
839 test_string = ''.join(map(compat_chr, range(len(example_sig))))
840 cache_res = func(test_string)
841 cache_spec = [ord(c) for c in cache_res]
842 expr_code = ' + '.join(gen_sig_code(cache_spec))
843 signature_id_tuple = '(%s)' % (
844 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
845 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
846 ' return %s\n') % (signature_id_tuple, expr_code)
847 self.to_screen('Extracted signature function:\n' + code)
849 def _parse_sig_js(self, jscode):
850 funcname = self._search_regex(
851 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
852 'Initial JS player signature function name')
854 jsi = JSInterpreter(jscode)
855 initial_function = jsi.extract_function(funcname)
856 return lambda s: initial_function([s])
858 def _parse_sig_swf(self, file_contents):
859 swfi = SWFInterpreter(file_contents)
860 TARGET_CLASSNAME = 'SignatureDecipher'
861 searched_class = swfi.extract_class(TARGET_CLASSNAME)
862 initial_function = swfi.extract_function(searched_class, 'decipher')
863 return lambda s: initial_function([s])
865 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
866 """Turn the encrypted s field into a working signature"""
868 if player_url is None:
869 raise ExtractorError('Cannot decrypt signature without player_url')
871 if player_url.startswith('//'):
872 player_url = 'https:' + player_url
874 player_id = (player_url, self._signature_cache_id(s))
875 if player_id not in self._player_cache:
876 func = self._extract_signature_function(
877 video_id, player_url, s
879 self._player_cache[player_id] = func
880 func = self._player_cache[player_id]
881 if self._downloader.params.get('youtube_print_sig_code'):
882 self._print_sig_code(func, s)
884 except Exception as e:
885 tb = traceback.format_exc()
886 raise ExtractorError(
887 'Signature extraction failed: ' + tb, cause=e)
889 def _get_subtitles(self, video_id, webpage):
891 subs_doc = self._download_xml(
892 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
893 video_id, note=False)
894 except ExtractorError as err:
895 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
899 for track in subs_doc.findall('track'):
900 lang = track.attrib['lang_code']
901 if lang in sub_lang_list:
904 for ext in ['sbv', 'vtt', 'srt']:
905 params = compat_urllib_parse.urlencode({
909 'name': track.attrib['name'].encode('utf-8'),
912 'url': 'https://www.youtube.com/api/timedtext?' + params,
915 sub_lang_list[lang] = sub_formats
916 if not sub_lang_list:
917 self._downloader.report_warning('video doesn\'t have subtitles')
921 def _get_ytplayer_config(self, video_id, webpage):
923 # User data may contain arbitrary character sequences that may affect
924 # JSON extraction with regex, e.g. when '};' is contained the second
925 # regex won't capture the whole JSON. Yet working around by trying more
926 # concrete regex first keeping in mind proper quoted string handling
927 # to be implemented in future that will replace this workaround (see
928 # https://github.com/rg3/youtube-dl/issues/7468,
929 # https://github.com/rg3/youtube-dl/pull/7599)
930 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
931 r';ytplayer\.config\s*=\s*({.+?});',
933 config = self._search_regex(
934 patterns, webpage, 'ytplayer.config', default=None)
936 return self._parse_json(
937 uppercase_escape(config), video_id, fatal=False)
939 def _get_automatic_captions(self, video_id, webpage):
940 """We need the webpage for getting the captions url, pass it as an
941 argument to speed up the process."""
942 self.to_screen('%s: Looking for automatic captions' % video_id)
943 player_config = self._get_ytplayer_config(video_id, webpage)
944 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
945 if not player_config:
946 self._downloader.report_warning(err_msg)
949 args = player_config['args']
950 caption_url = args['ttsurl']
951 timestamp = args['timestamp']
952 # We get the available subtitles
953 list_params = compat_urllib_parse.urlencode({
958 list_url = caption_url + '&' + list_params
959 caption_list = self._download_xml(list_url, video_id)
960 original_lang_node = caption_list.find('track')
961 if original_lang_node is None:
962 self._downloader.report_warning('Video doesn\'t have automatic captions')
964 original_lang = original_lang_node.attrib['lang_code']
965 caption_kind = original_lang_node.attrib.get('kind', '')
968 for lang_node in caption_list.findall('target'):
969 sub_lang = lang_node.attrib['lang_code']
971 for ext in ['sbv', 'vtt', 'srt']:
972 params = compat_urllib_parse.urlencode({
973 'lang': original_lang,
977 'kind': caption_kind,
980 'url': caption_url + '&' + params,
983 sub_lang_list[sub_lang] = sub_formats
985 # An extractor error can be raise by the download process if there are
986 # no automatic captions but there are subtitles
987 except (KeyError, ExtractorError):
988 self._downloader.report_warning(err_msg)
992 def extract_id(cls, url):
993 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
995 raise ExtractorError('Invalid URL: %s' % url)
996 video_id = mobj.group(2)
999 def _extract_from_m3u8(self, manifest_url, video_id):
1002 def _get_urls(_manifest):
1003 lines = _manifest.split('\n')
1004 urls = filter(lambda l: l and not l.startswith('#'),
1007 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1008 formats_urls = _get_urls(manifest)
1009 for format_url in formats_urls:
1010 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1011 url_map[itag] = format_url
1014 def _extract_annotations(self, video_id):
1015 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1016 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1018 def _parse_dash_manifest(
1019 self, video_id, dash_manifest_url, player_url, age_gate, fatal=True):
1020 def decrypt_sig(mobj):
1022 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1023 return '/signature/%s' % dec_s
1024 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
1025 dash_doc = self._download_xml(
1026 dash_manifest_url, video_id,
1027 note='Downloading DASH manifest',
1028 errnote='Could not download DASH manifest',
1031 if dash_doc is False:
1035 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
1036 mime_type = a.attrib.get('mimeType')
1037 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
1038 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
1041 if mime_type == 'text/vtt':
1042 # TODO implement WebVTT downloading
1044 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
1045 segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
1046 format_id = r.attrib['id']
1047 video_url = url_el.text
1048 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
1050 'format_id': format_id,
1052 'width': int_or_none(r.attrib.get('width')),
1053 'height': int_or_none(r.attrib.get('height')),
1054 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
1055 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
1056 'filesize': filesize,
1057 'fps': int_or_none(r.attrib.get('frameRate')),
1059 if segment_list is not None:
1061 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
1062 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
1063 'protocol': 'http_dash_segments',
1066 existing_format = next(
1067 fo for fo in formats
1068 if fo['format_id'] == format_id)
1069 except StopIteration:
1070 full_info = self._formats.get(format_id, {}).copy()
1072 codecs = r.attrib.get('codecs')
1074 if full_info.get('acodec') == 'none' and 'vcodec' not in full_info:
1075 full_info['vcodec'] = codecs
1076 elif full_info.get('vcodec') == 'none' and 'acodec' not in full_info:
1077 full_info['acodec'] = codecs
1078 formats.append(full_info)
1080 existing_format.update(f)
1082 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
1085 def _real_extract(self, url):
1086 url, smuggled_data = unsmuggle_url(url, {})
1089 'http' if self._downloader.params.get('prefer_insecure', False)
1094 parsed_url = compat_urllib_parse_urlparse(url)
1095 for component in [parsed_url.fragment, parsed_url.query]:
1096 query = compat_parse_qs(component)
1097 if start_time is None and 't' in query:
1098 start_time = parse_duration(query['t'][0])
1099 if start_time is None and 'start' in query:
1100 start_time = parse_duration(query['start'][0])
1101 if end_time is None and 'end' in query:
1102 end_time = parse_duration(query['end'][0])
1104 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1105 mobj = re.search(self._NEXT_URL_RE, url)
1107 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1108 video_id = self.extract_id(url)
1111 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1112 video_webpage = self._download_webpage(url, video_id)
1114 # Attempt to extract SWF player URL
1115 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1116 if mobj is not None:
1117 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1123 def add_dash_mpd(video_info):
1124 dash_mpd = video_info.get('dashmpd')
1125 if dash_mpd and dash_mpd[0] not in dash_mpds:
1126 dash_mpds.append(dash_mpd[0])
1129 embed_webpage = None
1131 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1133 # We simulate the access to the video from www.youtube.com/v/{video_id}
1134 # this can be viewed without login into Youtube
1135 url = proto + '://www.youtube.com/embed/%s' % video_id
1136 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1137 data = compat_urllib_parse.urlencode({
1138 'video_id': video_id,
1139 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1140 'sts': self._search_regex(
1141 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1143 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1144 video_info_webpage = self._download_webpage(
1145 video_info_url, video_id,
1146 note='Refetching age-gated info webpage',
1147 errnote='unable to download video info webpage')
1148 video_info = compat_parse_qs(video_info_webpage)
1149 add_dash_mpd(video_info)
1153 # Try looking directly into the video webpage
1154 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1156 args = ytplayer_config['args']
1157 if args.get('url_encoded_fmt_stream_map'):
1158 # Convert to the same format returned by compat_parse_qs
1159 video_info = dict((k, [v]) for k, v in args.items())
1160 add_dash_mpd(video_info)
1161 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1163 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1164 # We also try looking in get_video_info since it may contain different dashmpd
1165 # URL that points to a DASH manifest with possibly different itag set (some itags
1166 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1167 # manifest pointed by get_video_info's dashmpd).
1168 # The general idea is to take a union of itags of both DASH manifests (for example
1169 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1170 self.report_video_info_webpage_download(video_id)
1171 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1173 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1174 % (proto, video_id, el_type))
1175 video_info_webpage = self._download_webpage(
1177 video_id, note=False,
1178 errnote='unable to download video info webpage')
1179 get_video_info = compat_parse_qs(video_info_webpage)
1180 if get_video_info.get('use_cipher_signature') != ['True']:
1181 add_dash_mpd(get_video_info)
1183 video_info = get_video_info
1184 if 'token' in get_video_info:
1185 # Different get_video_info requests may report different results, e.g.
1186 # some may report video unavailability, but some may serve it without
1187 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1188 # the original webpage as well as el=info and el=embedded get_video_info
1189 # requests report video unavailability due to geo restriction while
1190 # el=detailpage succeeds and returns valid data). This is probably
1191 # due to YouTube measures against IP ranges of hosting providers.
1192 # Working around by preferring the first succeeded video_info containing
1193 # the token if no such video_info yet was found.
1194 if 'token' not in video_info:
1195 video_info = get_video_info
1197 if 'token' not in video_info:
1198 if 'reason' in video_info:
1199 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1200 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1202 raise ExtractorError('YouTube said: This video is available in %s only' % (
1203 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1205 raise ExtractorError(
1206 'YouTube said: %s' % video_info['reason'][0],
1207 expected=True, video_id=video_id)
1209 raise ExtractorError(
1210 '"token" parameter not in video info for unknown reason',
1214 if 'title' in video_info:
1215 video_title = video_info['title'][0]
1217 self._downloader.report_warning('Unable to extract video title')
1221 video_description = get_element_by_id("eow-description", video_webpage)
1222 if video_description:
1223 video_description = re.sub(r'''(?x)
1225 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1227 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1228 class="yt-uix-redirect-link"\s*>
1231 ''', r'\1', video_description)
1232 video_description = clean_html(video_description)
1234 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1236 video_description = unescapeHTML(fd_mobj.group(1))
1238 video_description = ''
1240 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1241 if not self._downloader.params.get('noplaylist'):
1244 multifeed_metadata_list = compat_urllib_parse_unquote_plus(video_info['multifeed_metadata_list'][0])
1245 for feed in multifeed_metadata_list.split(','):
1246 feed_data = compat_parse_qs(feed)
1248 '_type': 'url_transparent',
1249 'ie_key': 'Youtube',
1251 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1252 {'force_singlefeed': True}),
1253 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1255 feed_ids.append(feed_data['id'][0])
1257 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1258 % (', '.join(feed_ids), video_id))
1259 return self.playlist_result(entries, video_id, video_title, video_description)
1260 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1262 if 'view_count' in video_info:
1263 view_count = int(video_info['view_count'][0])
1267 # Check for "rental" videos
1268 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1269 raise ExtractorError('"rental" videos not supported')
1271 # Start extracting information
1272 self.report_information_extraction(video_id)
1275 if 'author' not in video_info:
1276 raise ExtractorError('Unable to extract uploader name')
1277 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1280 video_uploader_id = None
1281 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1282 if mobj is not None:
1283 video_uploader_id = mobj.group(1)
1285 self._downloader.report_warning('unable to extract uploader nickname')
1288 # We try first to get a high quality image:
1289 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1290 video_webpage, re.DOTALL)
1291 if m_thumb is not None:
1292 video_thumbnail = m_thumb.group(1)
1293 elif 'thumbnail_url' not in video_info:
1294 self._downloader.report_warning('unable to extract video thumbnail')
1295 video_thumbnail = None
1296 else: # don't panic if we can't find it
1297 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1300 upload_date = self._html_search_meta(
1301 'datePublished', video_webpage, 'upload date', default=None)
1303 upload_date = self._search_regex(
1304 [r'(?s)id="eow-date.*?>(.*?)</span>',
1305 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1306 video_webpage, 'upload date', default=None)
1308 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1309 upload_date = unified_strdate(upload_date)
1311 m_cat_container = self._search_regex(
1312 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1313 video_webpage, 'categories', default=None)
1315 category = self._html_search_regex(
1316 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1318 video_categories = None if category is None else [category]
1320 video_categories = None
1323 unescapeHTML(m.group('content'))
1324 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1326 def _extract_count(count_name):
1327 return str_to_int(self._search_regex(
1328 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1329 % re.escape(count_name),
1330 video_webpage, count_name, default=None))
1332 like_count = _extract_count('like')
1333 dislike_count = _extract_count('dislike')
1336 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1337 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1339 if 'length_seconds' not in video_info:
1340 self._downloader.report_warning('unable to extract video duration')
1341 video_duration = None
1343 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1346 video_annotations = None
1347 if self._downloader.params.get('writeannotations', False):
1348 video_annotations = self._extract_annotations(video_id)
1350 def _map_to_format_list(urlmap):
1352 for itag, video_real_url in urlmap.items():
1355 'url': video_real_url,
1356 'player_url': player_url,
1358 if itag in self._formats:
1359 dct.update(self._formats[itag])
1363 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1364 self.report_rtmp_download()
1366 'format_id': '_rtmp',
1368 'url': video_info['conn'][0],
1369 'player_url': player_url,
1371 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1372 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1373 if 'rtmpe%3Dyes' in encoded_url_map:
1374 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1376 for url_data_str in encoded_url_map.split(','):
1377 url_data = compat_parse_qs(url_data_str)
1378 if 'itag' not in url_data or 'url' not in url_data:
1380 format_id = url_data['itag'][0]
1381 url = url_data['url'][0]
1383 if 'sig' in url_data:
1384 url += '&signature=' + url_data['sig'][0]
1385 elif 's' in url_data:
1386 encrypted_sig = url_data['s'][0]
1387 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1389 jsplayer_url_json = self._search_regex(
1391 embed_webpage if age_gate else video_webpage,
1392 'JS player URL (1)', default=None)
1393 if not jsplayer_url_json and not age_gate:
1394 # We need the embed website after all
1395 if embed_webpage is None:
1396 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1397 embed_webpage = self._download_webpage(
1398 embed_url, video_id, 'Downloading embed webpage')
1399 jsplayer_url_json = self._search_regex(
1400 ASSETS_RE, embed_webpage, 'JS player URL')
1402 player_url = json.loads(jsplayer_url_json)
1403 if player_url is None:
1404 player_url_json = self._search_regex(
1405 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1406 video_webpage, 'age gate player URL')
1407 player_url = json.loads(player_url_json)
1409 if self._downloader.params.get('verbose'):
1410 if player_url is None:
1411 player_version = 'unknown'
1412 player_desc = 'unknown'
1414 if player_url.endswith('swf'):
1415 player_version = self._search_regex(
1416 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1417 'flash player', fatal=False)
1418 player_desc = 'flash player %s' % player_version
1420 player_version = self._search_regex(
1421 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1423 'html5 player', fatal=False)
1424 player_desc = 'html5 player %s' % player_version
1426 parts_sizes = self._signature_cache_id(encrypted_sig)
1427 self.to_screen('{%s} signature length %s, %s' %
1428 (format_id, parts_sizes, player_desc))
1430 signature = self._decrypt_signature(
1431 encrypted_sig, video_id, player_url, age_gate)
1432 url += '&signature=' + signature
1433 if 'ratebypass' not in url:
1434 url += '&ratebypass=yes'
1436 # Some itags are not included in DASH manifest thus corresponding formats will
1437 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1438 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1439 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1440 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1442 'format_id': format_id,
1444 'player_url': player_url,
1445 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1446 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1449 'fps': int_or_none(url_data.get('fps', [None])[0]),
1450 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1452 type_ = url_data.get('type', [None])[0]
1454 type_split = type_.split(';')
1455 kind_ext = type_split[0].split('/')
1456 if len(kind_ext) == 2:
1457 kind, ext = kind_ext
1459 if kind in ('audio', 'video'):
1461 for mobj in re.finditer(
1462 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1463 if mobj.group('key') == 'codecs':
1464 codecs = mobj.group('val')
1467 codecs = codecs.split(',')
1468 if len(codecs) == 2:
1469 acodec, vcodec = codecs[0], codecs[1]
1471 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1476 if format_id in self._formats:
1477 dct.update(self._formats[format_id])
1479 elif video_info.get('hlsvp'):
1480 manifest_url = video_info['hlsvp'][0]
1481 url_map = self._extract_from_m3u8(manifest_url, video_id)
1482 formats = _map_to_format_list(url_map)
1483 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1484 for a_format in formats:
1485 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1487 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1489 # Look for the DASH manifest
1490 if self._downloader.params.get('youtube_include_dash_manifest', True):
1491 dash_mpd_fatal = True
1492 for dash_manifest_url in dash_mpds:
1495 for df in self._parse_dash_manifest(
1496 video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
1497 # Do not overwrite DASH format found in some previous DASH manifest
1498 if df['format_id'] not in dash_formats:
1499 dash_formats[df['format_id']] = df
1500 # Additional DASH manifests may end up in HTTP Error 403 therefore
1501 # allow them to fail without bug report message if we already have
1502 # some DASH manifest succeeded. This is temporary workaround to reduce
1503 # burst of bug reports until we figure out the reason and whether it
1504 # can be fixed at all.
1505 dash_mpd_fatal = False
1506 except (ExtractorError, KeyError) as e:
1507 self.report_warning(
1508 'Skipping DASH manifest: %r' % e, video_id)
1510 # Remove the formats we found through non-DASH, they
1511 # contain less info and it can be wrong, because we use
1512 # fixed values (for example the resolution). See
1513 # https://github.com/rg3/youtube-dl/issues/5774 for an
1515 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1516 formats.extend(dash_formats.values())
1518 # Check for malformed aspect ratio
1519 stretched_m = re.search(
1520 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1523 w = float(stretched_m.group('w'))
1524 h = float(stretched_m.group('h'))
1525 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1526 # We will only process correct ratios.
1530 if f.get('vcodec') != 'none':
1531 f['stretched_ratio'] = ratio
1533 self._sort_formats(formats)
1537 'uploader': video_uploader,
1538 'uploader_id': video_uploader_id,
1539 'upload_date': upload_date,
1540 'title': video_title,
1541 'thumbnail': video_thumbnail,
1542 'description': video_description,
1543 'categories': video_categories,
1545 'subtitles': video_subtitles,
1546 'automatic_captions': automatic_captions,
1547 'duration': video_duration,
1548 'age_limit': 18 if age_gate else 0,
1549 'annotations': video_annotations,
1550 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1551 'view_count': view_count,
1552 'like_count': like_count,
1553 'dislike_count': dislike_count,
1554 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1557 'start_time': start_time,
1558 'end_time': end_time,
1562 class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor):
1563 IE_DESC = 'YouTube.com playlists'
1564 _VALID_URL = r"""(?x)(?:
1569 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1570 \? (?:.*?[&;])*? (?:p|a|list)=
1574 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1575 # Top tracks, they can also include dots
1580 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1582 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1583 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1584 IE_NAME = 'youtube:playlist'
1586 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1588 'title': 'ytdl test PL',
1589 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1591 'playlist_count': 3,
1593 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1595 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1596 'title': 'YDL_Empty_List',
1598 'playlist_count': 0,
1600 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1601 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1603 'title': '29C3: Not my department',
1604 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1606 'playlist_count': 95,
1608 'note': 'issue #673',
1609 'url': 'PLBB231211A4F62143',
1611 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1612 'id': 'PLBB231211A4F62143',
1614 'playlist_mincount': 26,
1616 'note': 'Large playlist',
1617 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1619 'title': 'Uploads from Cauchemar',
1620 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1622 'playlist_mincount': 799,
1624 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1626 'title': 'YDL_safe_search',
1627 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1629 'playlist_count': 2,
1632 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1633 'playlist_count': 4,
1636 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1639 'note': 'Embedded SWF player',
1640 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1641 'playlist_count': 4,
1644 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1647 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1648 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1650 'title': 'Uploads from Interstellar Movie',
1651 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1653 'playlist_mincout': 21,
1656 def _real_initialize(self):
1659 def _extract_mix(self, playlist_id):
1660 # The mixes are generated from a single video
1661 # the id of the playlist is just 'RD' + video_id
1662 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1663 webpage = self._download_webpage(
1664 url, playlist_id, 'Downloading Youtube mix')
1665 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1667 search_title('playlist-title') or
1668 search_title('title long-title') or
1669 search_title('title'))
1670 title = clean_html(title_span)
1671 ids = orderedSet(re.findall(
1672 r'''(?xs)data-video-username=".*?".*?
1673 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1675 url_results = self._ids_to_results(ids)
1677 return self.playlist_result(url_results, playlist_id, title)
1679 def _extract_playlist(self, playlist_id):
1680 url = self._TEMPLATE_URL % playlist_id
1681 page = self._download_webpage(url, playlist_id)
1683 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1684 match = match.strip()
1685 # Check if the playlist exists or is private
1686 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1687 raise ExtractorError(
1688 'The playlist doesn\'t exist or is private, use --username or '
1689 '--netrc to access it.',
1691 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1692 raise ExtractorError(
1693 'Invalid parameters. Maybe URL is incorrect.',
1695 elif re.match(r'[^<]*Choose your language[^<]*', match):
1698 self.report_warning('Youtube gives an alert message: ' + match)
1700 playlist_title = self._html_search_regex(
1701 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1704 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1706 def _real_extract(self, url):
1707 # Extract playlist id
1708 mobj = re.match(self._VALID_URL, url)
1710 raise ExtractorError('Invalid URL: %s' % url)
1711 playlist_id = mobj.group(1) or mobj.group(2)
1713 # Check if it's a video-specific URL
1714 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1715 if 'v' in query_dict:
1716 video_id = query_dict['v'][0]
1717 if self._downloader.params.get('noplaylist'):
1718 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1719 return self.url_result(video_id, 'Youtube', video_id=video_id)
1721 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1723 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1724 # Mixes require a custom extraction process
1725 return self._extract_mix(playlist_id)
1727 return self._extract_playlist(playlist_id)
1730 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1731 IE_DESC = 'YouTube.com channels'
1732 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1733 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1734 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1735 IE_NAME = 'youtube:channel'
1737 'note': 'paginated channel',
1738 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1739 'playlist_mincount': 91,
1741 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1742 'title': 'Uploads from lex will',
1745 'note': 'Age restricted channel',
1746 # from https://www.youtube.com/user/DeusExOfficial
1747 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1748 'playlist_mincount': 64,
1750 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1751 'title': 'Uploads from Deus Ex',
1755 def _real_extract(self, url):
1756 channel_id = self._match_id(url)
1758 url = self._TEMPLATE_URL % channel_id
1760 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1761 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1762 # otherwise fallback on channel by page extraction
1763 channel_page = self._download_webpage(
1764 url + '?view=57', channel_id,
1765 'Downloading channel page', fatal=False)
1766 if channel_page is False:
1767 channel_playlist_id = False
1769 channel_playlist_id = self._html_search_meta(
1770 'channelId', channel_page, 'channel id', default=None)
1771 if not channel_playlist_id:
1772 channel_playlist_id = self._search_regex(
1773 r'data-(?:channel-external-|yt)id="([^"]+)"',
1774 channel_page, 'channel id', default=None)
1775 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1776 playlist_id = 'UU' + channel_playlist_id[2:]
1777 return self.url_result(
1778 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1780 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1781 autogenerated = re.search(r'''(?x)
1783 channel-header-autogenerated-label|
1784 yt-channel-title-autogenerated
1785 )[^"]*"''', channel_page) is not None
1788 # The videos are contained in a single page
1789 # the ajax pages can't be used, they are empty
1792 video_id, 'Youtube', video_id=video_id,
1793 video_title=video_title)
1794 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1795 return self.playlist_result(entries, channel_id)
1797 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1800 class YoutubeUserIE(YoutubeChannelIE):
1801 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1802 _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_-]+)'
1803 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1804 IE_NAME = 'youtube:user'
1807 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1808 'playlist_mincount': 320,
1810 'title': 'TheLinuxFoundation',
1813 'url': 'ytuser:phihag',
1814 'only_matching': True,
1818 def suitable(cls, url):
1819 # Don't return True if the url can be extracted with other youtube
1820 # extractor, the regex would is too permissive and it would match.
1821 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1822 if any(ie.suitable(url) for ie in other_ies):
1825 return super(YoutubeUserIE, cls).suitable(url)
1828 class YoutubeUserPlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1829 IE_DESC = 'YouTube.com user playlists'
1830 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/user/(?P<id>[^/]+)/playlists'
1831 IE_NAME = 'youtube:user:playlists'
1834 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1835 'playlist_mincount': 4,
1837 'id': 'ThirstForScience',
1838 'title': 'Thirst for Science',
1841 # with "Load more" button
1842 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1843 'playlist_mincount': 70,
1846 'title': 'Игорь Клейнер',
1851 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1852 IE_DESC = 'YouTube.com searches'
1853 # there doesn't appear to be a real limit, for example if you search for
1854 # 'python' you get more than 8.000.000 results
1855 _MAX_RESULTS = float('inf')
1856 IE_NAME = 'youtube:search'
1857 _SEARCH_KEY = 'ytsearch'
1858 _EXTRA_QUERY_ARGS = {}
1861 def _get_n_results(self, query, n):
1862 """Get a specified number of results for a query"""
1867 for pagenum in itertools.count(1):
1869 'search_query': query.encode('utf-8'),
1873 url_query.update(self._EXTRA_QUERY_ARGS)
1874 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1875 data = self._download_json(
1876 result_url, video_id='query "%s"' % query,
1877 note='Downloading page %s' % pagenum,
1878 errnote='Unable to download API page')
1879 html_content = data[1]['body']['content']
1881 if 'class="search-message' in html_content:
1882 raise ExtractorError(
1883 '[youtube] No video results', expected=True)
1885 new_videos = self._ids_to_results(orderedSet(re.findall(
1886 r'href="/watch\?v=(.{11})', html_content)))
1887 videos += new_videos
1888 if not new_videos or len(videos) > limit:
1893 return self.playlist_result(videos, query)
1896 class YoutubeSearchDateIE(YoutubeSearchIE):
1897 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1898 _SEARCH_KEY = 'ytsearchdate'
1899 IE_DESC = 'YouTube.com searches, newest videos first'
1900 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1903 class YoutubeSearchURLIE(InfoExtractor):
1904 IE_DESC = 'YouTube.com search URLs'
1905 IE_NAME = 'youtube:search_url'
1906 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1908 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1909 'playlist_mincount': 5,
1911 'title': 'youtube-dl test video',
1915 def _real_extract(self, url):
1916 mobj = re.match(self._VALID_URL, url)
1917 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
1919 webpage = self._download_webpage(url, query)
1920 result_code = self._search_regex(
1921 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1923 part_codes = re.findall(
1924 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
1926 for part_code in part_codes:
1927 part_title = self._html_search_regex(
1928 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1929 part_url_snippet = self._html_search_regex(
1930 r'(?s)href="([^"]+)"', part_code, 'item URL')
1931 part_url = compat_urlparse.urljoin(
1932 'https://www.youtube.com/', part_url_snippet)
1936 'title': part_title,
1940 '_type': 'playlist',
1946 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
1947 IE_DESC = 'YouTube.com (multi-season) shows'
1948 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1949 IE_NAME = 'youtube:show'
1951 'url': 'https://www.youtube.com/show/airdisasters',
1952 'playlist_mincount': 5,
1954 'id': 'airdisasters',
1955 'title': 'Air Disasters',
1959 def _real_extract(self, url):
1960 playlist_id = self._match_id(url)
1961 return super(YoutubeShowIE, self)._real_extract(
1962 'https://www.youtube.com/show/%s/playlists' % playlist_id)
1965 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1967 Base class for feed extractors
1968 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1970 _LOGIN_REQUIRED = True
1974 return 'youtube:%s' % self._FEED_NAME
1976 def _real_initialize(self):
1979 def _real_extract(self, url):
1980 page = self._download_webpage(
1981 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1983 # The extraction process is the same as for playlists, but the regex
1984 # for the video ids doesn't contain an index
1986 more_widget_html = content_html = page
1987 for page_num in itertools.count(1):
1988 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1990 # 'recommended' feed has infinite 'load more' and each new portion spins
1991 # the same videos in (sometimes) slightly different order, so we'll check
1992 # for unicity and break when portion has no new videos
1993 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1999 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2003 more = self._download_json(
2004 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2005 'Downloading page #%s' % page_num,
2006 transform_source=uppercase_escape)
2007 content_html = more['content_html']
2008 more_widget_html = more['load_more_widget_html']
2010 return self.playlist_result(
2011 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2014 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2015 IE_NAME = 'youtube:watchlater'
2016 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2017 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
2019 _TESTS = [] # override PlaylistIE tests
2021 def _real_extract(self, url):
2022 return self._extract_playlist('WL')
2025 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2026 IE_NAME = 'youtube:favorites'
2027 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2028 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2029 _LOGIN_REQUIRED = True
2031 def _real_extract(self, url):
2032 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2033 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2034 return self.url_result(playlist_id, 'YoutubePlaylist')
2037 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2038 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2039 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2040 _FEED_NAME = 'recommended'
2041 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2044 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2045 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2046 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2047 _FEED_NAME = 'subscriptions'
2048 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2051 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2052 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2053 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2054 _FEED_NAME = 'history'
2055 _PLAYLIST_TITLE = 'Youtube History'
2058 class YoutubeTruncatedURLIE(InfoExtractor):
2059 IE_NAME = 'youtube:truncated_url'
2060 IE_DESC = False # Do not list
2061 _VALID_URL = r'''(?x)
2063 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2066 annotation_id=annotation_[^&]+|
2072 attribution_link\?a=[^&]+
2078 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2079 'only_matching': True,
2081 'url': 'http://www.youtube.com/watch?',
2082 'only_matching': True,
2084 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2085 'only_matching': True,
2087 'url': 'https://www.youtube.com/watch?feature=foo',
2088 'only_matching': True,
2090 'url': 'https://www.youtube.com/watch?hl=en-GB',
2091 'only_matching': True,
2093 'url': 'https://www.youtube.com/watch?t=2372',
2094 'only_matching': True,
2097 def _real_extract(self, url):
2098 raise ExtractorError(
2099 'Did you forget to quote the URL? Remember that & is a meta '
2100 'character in most shells, so you want to put the URL in quotes, '
2102 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2103 ' or simply youtube-dl BaW_jenozKc .',
2107 class YoutubeTruncatedIDIE(InfoExtractor):
2108 IE_NAME = 'youtube:truncated_id'
2109 IE_DESC = False # Do not list
2110 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2113 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2114 'only_matching': True,
2117 def _real_extract(self, url):
2118 video_id = self._match_id(url)
2119 raise ExtractorError(
2120 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),