3 from __future__ import unicode_literals
13 from .common import InfoExtractor, SearchInfoExtractor
14 from ..jsinterp import JSInterpreter
15 from ..swfinterp import SWFInterpreter
16 from ..compat import (
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlparse,
32 get_element_by_attribute,
50 class YoutubeBaseInfoExtractor(InfoExtractor):
51 """Provide base functions for Youtube extractors"""
52 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
53 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
54 _NETRC_MACHINE = 'youtube'
55 # If True it will raise an error if no login info is provided
56 _LOGIN_REQUIRED = False
58 def _set_language(self):
60 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
61 # YouTube sets the expire time to about two months
62 expire_time=time.time() + 2 * 30 * 24 * 3600)
64 def _ids_to_results(self, ids):
66 self.url_result(vid_id, 'Youtube', video_id=vid_id)
71 Attempt to log in to YouTube.
72 True is returned if successful or skipped.
73 False is returned if login failed.
75 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
77 (username, password) = self._get_login_info()
78 # No authentication to be performed
80 if self._LOGIN_REQUIRED:
81 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
84 login_page = self._download_webpage(
85 self._LOGIN_URL, None,
86 note='Downloading login page',
87 errnote='unable to fetch login page', fatal=False)
88 if login_page is False:
91 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
92 login_page, 'Login GALX parameter')
96 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
101 'PersistentCookie': 'yes',
103 'bgresponse': 'js_disabled',
104 'checkConnection': '',
105 'checkedDomains': 'youtube',
112 'service': 'youtube',
117 login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii')
119 req = sanitized_Request(self._LOGIN_URL, login_data)
120 login_results = self._download_webpage(
122 note='Logging in', errnote='unable to log in', fatal=False)
123 if login_results is False:
126 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
127 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
130 # TODO add SMS and phone call support - these require making a request and then prompting the user
132 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
133 tfa_code = self._get_tfa_info('2-step verification code')
136 self._downloader.report_warning(
137 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
138 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
141 tfa_code = remove_start(tfa_code, 'G-')
143 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
145 tfa_form_strs.update({
150 tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii')
152 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
153 tfa_results = self._download_webpage(
155 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
157 if tfa_results is False:
160 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
161 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
163 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
164 self._downloader.report_warning('unable to log in - did the page structure change?')
166 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
167 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
170 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
171 self._downloader.report_warning('unable to log in: bad username or password')
175 def _real_initialize(self):
176 if self._downloader is None:
179 if not self._login():
183 class YoutubeEntryListBaseInfoExtractor(InfoExtractor):
184 # Extract entries from page with "Load more" button
185 def _entries(self, page, playlist_id):
186 more_widget_html = content_html = page
187 for page_num in itertools.count(1):
188 for entry in self._process_page(content_html):
191 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
195 more = self._download_json(
196 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
197 'Downloading page #%s' % page_num,
198 transform_source=uppercase_escape)
199 content_html = more['content_html']
200 if not content_html.strip():
201 # Some webpages show a "Load more" button but they don't
204 more_widget_html = more['load_more_widget_html']
207 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
208 def _process_page(self, content):
209 for video_id, video_title in self.extract_videos_from_page(content):
210 yield self.url_result(video_id, 'Youtube', video_id, video_title)
212 def extract_videos_from_page(self, page):
215 for mobj in re.finditer(self._VIDEO_RE, page):
216 # The link with index 0 is not the first video of the playlist (not sure if still actual)
217 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
219 video_id = mobj.group('id')
220 video_title = unescapeHTML(mobj.group('title'))
222 video_title = video_title.strip()
224 idx = ids_in_page.index(video_id)
225 if video_title and not titles_in_page[idx]:
226 titles_in_page[idx] = video_title
228 ids_in_page.append(video_id)
229 titles_in_page.append(video_title)
230 return zip(ids_in_page, titles_in_page)
233 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
234 def _process_page(self, content):
235 for playlist_id in re.findall(r'href="/?playlist\?list=(.+?)"', content):
236 yield self.url_result(
237 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
239 def _real_extract(self, url):
240 playlist_id = self._match_id(url)
241 webpage = self._download_webpage(url, playlist_id)
242 title = self._og_search_title(webpage, fatal=False)
243 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
246 class YoutubeIE(YoutubeBaseInfoExtractor):
247 IE_DESC = 'YouTube.com'
248 _VALID_URL = r"""(?x)^
250 (?:https?://|//) # http(s):// or protocol-independent URL
251 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
252 (?:www\.)?deturl\.com/www\.youtube\.com/|
253 (?:www\.)?pwnyoutube\.com/|
254 (?:www\.)?yourepeat\.com/|
255 tube\.majestyc\.net/|
256 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
257 (?:.*?\#/)? # handle anchor (#/) redirect urls
258 (?: # the various things that can precede the ID:
259 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
260 |(?: # or the v= param in all its forms
261 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
262 (?:\?|\#!?) # the params delimiter ? or # or #!
263 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
268 youtu\.be| # just youtu.be/xxxx
269 vid\.plus # or vid.plus/xxxx
271 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
273 )? # all until now is optional -> you can pass the naked ID
274 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
275 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
276 (?(1).+)? # if we found the ID, everything can follow
278 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
280 '5': {'ext': 'flv', 'width': 400, 'height': 240},
281 '6': {'ext': 'flv', 'width': 450, 'height': 270},
282 '13': {'ext': '3gp'},
283 '17': {'ext': '3gp', 'width': 176, 'height': 144},
284 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
285 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
286 '34': {'ext': 'flv', 'width': 640, 'height': 360},
287 '35': {'ext': 'flv', 'width': 854, 'height': 480},
288 '36': {'ext': '3gp', 'width': 320, 'height': 240},
289 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
290 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
291 '43': {'ext': 'webm', 'width': 640, 'height': 360},
292 '44': {'ext': 'webm', 'width': 854, 'height': 480},
293 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
294 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
295 '59': {'ext': 'mp4', 'width': 854, 'height': 480},
296 '78': {'ext': 'mp4', 'width': 854, 'height': 480},
300 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
301 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
302 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
303 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
304 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
305 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
306 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
308 # Apple HTTP Live Streaming
309 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
310 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
311 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
312 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
313 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
314 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
315 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
318 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
319 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
320 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
321 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
322 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
323 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
324 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
325 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
326 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
327 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
328 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
331 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
332 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
333 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
336 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
337 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
338 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
339 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
340 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
341 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
342 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'vp9'},
343 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
344 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
345 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
346 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
347 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
348 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
349 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
350 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
351 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
352 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
353 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
354 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
355 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
356 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'vp9'},
357 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
360 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
361 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
363 # Dash webm audio with opus inside
364 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
365 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
366 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
369 '_rtmp': {'protocol': 'rtmp'},
375 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
379 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
380 'uploader': 'Philipp Hagemeister',
381 'uploader_id': 'phihag',
382 'upload_date': '20121002',
383 '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 .',
384 'categories': ['Science & Technology'],
385 'tags': ['youtube-dl'],
387 'dislike_count': int,
393 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
394 'note': 'Test generic use_cipher_signature video (#897)',
398 'upload_date': '20120506',
399 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
400 'alt_title': 'I Love It (feat. Charli XCX)',
401 'description': 'md5:782e8651347686cba06e58f71ab51773',
402 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
403 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
404 'iconic ep', 'iconic', 'love', 'it'],
405 'uploader': 'Icona Pop',
406 'uploader_id': 'IconaPop',
407 'creator': 'Icona Pop',
411 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
412 'note': 'Test VEVO video with age protection (#956)',
416 'upload_date': '20130703',
417 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
418 'alt_title': 'Tunnel Vision',
419 'description': 'md5:64249768eec3bc4276236606ea996373',
420 'uploader': 'justintimberlakeVEVO',
421 'uploader_id': 'justintimberlakeVEVO',
422 'creator': 'Justin Timberlake',
427 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
428 'note': 'Embed-only video (#1746)',
432 'upload_date': '20120608',
433 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
434 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
435 'uploader': 'SET India',
436 'uploader_id': 'setindia',
441 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY',
442 'note': 'Use the first video ID in the URL',
446 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
447 'uploader': 'Philipp Hagemeister',
448 'uploader_id': 'phihag',
449 'upload_date': '20121002',
450 '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 .',
451 'categories': ['Science & Technology'],
452 'tags': ['youtube-dl'],
454 'dislike_count': int,
457 'skip_download': True,
461 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
462 'note': '256k DASH audio (format 141) via DASH manifest',
466 'upload_date': '20121002',
467 'uploader_id': '8KVIDEO',
469 'uploader': '8KVIDEO',
470 'title': 'UHDTV TEST 8K VIDEO.mp4'
473 'youtube_include_dash_manifest': True,
477 # DASH manifest with encrypted signature
479 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
483 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
484 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
485 'uploader': 'AfrojackVEVO',
486 'uploader_id': 'AfrojackVEVO',
487 'upload_date': '20131011',
490 'youtube_include_dash_manifest': True,
494 # JS player signature function name containing $
496 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
500 'title': 'Taylor Swift - Shake It Off',
501 'alt_title': 'Shake It Off',
502 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
503 'uploader': 'TaylorSwiftVEVO',
504 'uploader_id': 'TaylorSwiftVEVO',
505 'upload_date': '20140818',
506 'creator': 'Taylor Swift',
509 'youtube_include_dash_manifest': True,
515 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
519 'upload_date': '20100909',
520 'uploader': 'The Amazing Atheist',
521 'uploader_id': 'TheAmazingAtheist',
522 'title': 'Burning Everyone\'s Koran',
523 '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',
526 # Normal age-gate video (No vevo, embed allowed)
528 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
532 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
533 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
534 'uploader': 'The Witcher',
535 'uploader_id': 'WitcherGame',
536 'upload_date': '20140605',
540 # Age-gate video with encrypted signature
542 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
546 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
547 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
548 'uploader': 'LloydVEVO',
549 'uploader_id': 'LloydVEVO',
550 'upload_date': '20110629',
554 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
556 'url': '__2ABJjxzNo',
560 'upload_date': '20100430',
561 'uploader_id': 'deadmau5',
562 'creator': 'deadmau5',
563 'description': 'md5:12c56784b8032162bb936a5f76d55360',
564 'uploader': 'deadmau5',
565 'title': 'Deadmau5 - Some Chords (HD)',
566 'alt_title': 'Some Chords',
568 'expected_warnings': [
569 'DASH manifest missing',
572 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
574 'url': 'lqQg6PlCWgI',
578 'upload_date': '20150827',
579 'uploader_id': 'olympic',
580 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
581 'uploader': 'Olympics',
582 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
585 'skip_download': 'requires avconv',
590 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
594 'stretched_ratio': 16 / 9.,
595 'upload_date': '20110310',
596 'uploader_id': 'AllenMeow',
597 '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',
599 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
602 # url_encoded_fmt_stream_map is empty string
604 'url': 'qEJwOuvDf7I',
608 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
610 'upload_date': '20150404',
611 'uploader_id': 'spbelect',
612 'uploader': 'Наблюдатели Петербурга',
615 'skip_download': 'requires avconv',
617 'skip': 'This live event has ended.',
619 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
621 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
625 'title': 'md5:7b81415841e02ecd4313668cde88737a',
626 'description': 'md5:116377fd2963b81ec4ce64b542173306',
627 'upload_date': '20150625',
628 'uploader_id': 'dorappi2000',
629 'uploader': 'dorappi2000',
630 'formats': 'mincount:33',
633 # DASH manifest with segment_list
635 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
636 'md5': '8ce563a1d667b599d21064e982ab9e31',
640 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
641 'uploader': 'Airtek',
642 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
643 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
644 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
647 'youtube_include_dash_manifest': True,
648 'format': '135', # bestvideo
652 # Multifeed videos (multiple cameras), URL is for Main Camera
653 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
656 'title': 'teamPGP: Rocket League Noob Stream',
657 'description': 'md5:dc7872fb300e143831327f1bae3af010',
663 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
664 'description': 'md5:dc7872fb300e143831327f1bae3af010',
665 'upload_date': '20150721',
666 'uploader': 'Beer Games Beer',
667 'uploader_id': 'beergamesbeer',
673 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
674 'description': 'md5:dc7872fb300e143831327f1bae3af010',
675 'upload_date': '20150721',
676 'uploader': 'Beer Games Beer',
677 'uploader_id': 'beergamesbeer',
683 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
684 'description': 'md5:dc7872fb300e143831327f1bae3af010',
685 'upload_date': '20150721',
686 'uploader': 'Beer Games Beer',
687 'uploader_id': 'beergamesbeer',
693 'title': 'teamPGP: Rocket League Noob Stream (zim)',
694 'description': 'md5:dc7872fb300e143831327f1bae3af010',
695 'upload_date': '20150721',
696 'uploader': 'Beer Games Beer',
697 'uploader_id': 'beergamesbeer',
701 'skip_download': True,
705 'url': 'http://vid.plus/FlRa-iH7PGw',
706 'only_matching': True,
709 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
710 # Also tests cut-off URL expansion in video description (see
711 # https://github.com/rg3/youtube-dl/issues/1892,
712 # https://github.com/rg3/youtube-dl/issues/8164)
713 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
717 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
718 'alt_title': 'Dark Walk',
719 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
720 'upload_date': '20151119',
721 'uploader_id': 'IronSoulElf',
722 'uploader': 'IronSoulElf',
723 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
726 'skip_download': True,
730 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
731 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
732 'only_matching': True,
735 # Video with yt:stretch=17:0
736 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
740 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
741 'description': 'md5:ee18a25c350637c8faff806845bddee9',
742 'upload_date': '20151107',
743 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
744 'uploader': 'CH GAMER DROID',
747 'skip_download': True,
751 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
752 'only_matching': True,
756 def __init__(self, *args, **kwargs):
757 super(YoutubeIE, self).__init__(*args, **kwargs)
758 self._player_cache = {}
760 def report_video_info_webpage_download(self, video_id):
761 """Report attempt to download video info webpage."""
762 self.to_screen('%s: Downloading video info webpage' % video_id)
764 def report_information_extraction(self, video_id):
765 """Report attempt to extract video information."""
766 self.to_screen('%s: Extracting video information' % video_id)
768 def report_unavailable_format(self, video_id, format):
769 """Report extracted video URL."""
770 self.to_screen('%s: Format %s not available' % (video_id, format))
772 def report_rtmp_download(self):
773 """Indicate the download will use the RTMP protocol."""
774 self.to_screen('RTMP download detected')
776 def _signature_cache_id(self, example_sig):
777 """ Return a string representation of a signature """
778 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
780 def _extract_signature_function(self, video_id, player_url, example_sig):
782 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
785 raise ExtractorError('Cannot identify player %r' % player_url)
786 player_type = id_m.group('ext')
787 player_id = id_m.group('id')
789 # Read from filesystem cache
790 func_id = '%s_%s_%s' % (
791 player_type, player_id, self._signature_cache_id(example_sig))
792 assert os.path.basename(func_id) == func_id
794 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
795 if cache_spec is not None:
796 return lambda s: ''.join(s[i] for i in cache_spec)
799 'Downloading player %s' % player_url
800 if self._downloader.params.get('verbose') else
801 'Downloading %s player %s' % (player_type, player_id)
803 if player_type == 'js':
804 code = self._download_webpage(
805 player_url, video_id,
807 errnote='Download of %s failed' % player_url)
808 res = self._parse_sig_js(code)
809 elif player_type == 'swf':
810 urlh = self._request_webpage(
811 player_url, video_id,
813 errnote='Download of %s failed' % player_url)
815 res = self._parse_sig_swf(code)
817 assert False, 'Invalid player type %r' % player_type
819 test_string = ''.join(map(compat_chr, range(len(example_sig))))
820 cache_res = res(test_string)
821 cache_spec = [ord(c) for c in cache_res]
823 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
826 def _print_sig_code(self, func, example_sig):
827 def gen_sig_code(idxs):
828 def _genslice(start, end, step):
829 starts = '' if start == 0 else str(start)
830 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
831 steps = '' if step == 1 else (':%d' % step)
832 return 's[%s%s%s]' % (starts, ends, steps)
835 # Quelch pyflakes warnings - start will be set when step is set
836 start = '(Never used)'
837 for i, prev in zip(idxs[1:], idxs[:-1]):
841 yield _genslice(start, prev, step)
844 if i - prev in [-1, 1]:
853 yield _genslice(start, i, step)
855 test_string = ''.join(map(compat_chr, range(len(example_sig))))
856 cache_res = func(test_string)
857 cache_spec = [ord(c) for c in cache_res]
858 expr_code = ' + '.join(gen_sig_code(cache_spec))
859 signature_id_tuple = '(%s)' % (
860 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
861 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
862 ' return %s\n') % (signature_id_tuple, expr_code)
863 self.to_screen('Extracted signature function:\n' + code)
865 def _parse_sig_js(self, jscode):
866 funcname = self._search_regex(
867 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
868 'Initial JS player signature function name')
870 jsi = JSInterpreter(jscode)
871 initial_function = jsi.extract_function(funcname)
872 return lambda s: initial_function([s])
874 def _parse_sig_swf(self, file_contents):
875 swfi = SWFInterpreter(file_contents)
876 TARGET_CLASSNAME = 'SignatureDecipher'
877 searched_class = swfi.extract_class(TARGET_CLASSNAME)
878 initial_function = swfi.extract_function(searched_class, 'decipher')
879 return lambda s: initial_function([s])
881 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
882 """Turn the encrypted s field into a working signature"""
884 if player_url is None:
885 raise ExtractorError('Cannot decrypt signature without player_url')
887 if player_url.startswith('//'):
888 player_url = 'https:' + player_url
890 player_id = (player_url, self._signature_cache_id(s))
891 if player_id not in self._player_cache:
892 func = self._extract_signature_function(
893 video_id, player_url, s
895 self._player_cache[player_id] = func
896 func = self._player_cache[player_id]
897 if self._downloader.params.get('youtube_print_sig_code'):
898 self._print_sig_code(func, s)
900 except Exception as e:
901 tb = traceback.format_exc()
902 raise ExtractorError(
903 'Signature extraction failed: ' + tb, cause=e)
905 def _get_subtitles(self, video_id, webpage):
907 subs_doc = self._download_xml(
908 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
909 video_id, note=False)
910 except ExtractorError as err:
911 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
915 for track in subs_doc.findall('track'):
916 lang = track.attrib['lang_code']
917 if lang in sub_lang_list:
920 for ext in ['sbv', 'vtt', 'srt']:
921 params = compat_urllib_parse.urlencode({
925 'name': track.attrib['name'].encode('utf-8'),
928 'url': 'https://www.youtube.com/api/timedtext?' + params,
931 sub_lang_list[lang] = sub_formats
932 if not sub_lang_list:
933 self._downloader.report_warning('video doesn\'t have subtitles')
937 def _get_ytplayer_config(self, video_id, webpage):
939 # User data may contain arbitrary character sequences that may affect
940 # JSON extraction with regex, e.g. when '};' is contained the second
941 # regex won't capture the whole JSON. Yet working around by trying more
942 # concrete regex first keeping in mind proper quoted string handling
943 # to be implemented in future that will replace this workaround (see
944 # https://github.com/rg3/youtube-dl/issues/7468,
945 # https://github.com/rg3/youtube-dl/pull/7599)
946 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
947 r';ytplayer\.config\s*=\s*({.+?});',
949 config = self._search_regex(
950 patterns, webpage, 'ytplayer.config', default=None)
952 return self._parse_json(
953 uppercase_escape(config), video_id, fatal=False)
955 def _get_automatic_captions(self, video_id, webpage):
956 """We need the webpage for getting the captions url, pass it as an
957 argument to speed up the process."""
958 self.to_screen('%s: Looking for automatic captions' % video_id)
959 player_config = self._get_ytplayer_config(video_id, webpage)
960 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
961 if not player_config:
962 self._downloader.report_warning(err_msg)
965 args = player_config['args']
966 caption_url = args['ttsurl']
968 self._downloader.report_warning(err_msg)
970 timestamp = args['timestamp']
971 # We get the available subtitles
972 list_params = compat_urllib_parse.urlencode({
977 list_url = caption_url + '&' + list_params
978 caption_list = self._download_xml(list_url, video_id)
979 original_lang_node = caption_list.find('track')
980 if original_lang_node is None:
981 self._downloader.report_warning('Video doesn\'t have automatic captions')
983 original_lang = original_lang_node.attrib['lang_code']
984 caption_kind = original_lang_node.attrib.get('kind', '')
987 for lang_node in caption_list.findall('target'):
988 sub_lang = lang_node.attrib['lang_code']
990 for ext in ['sbv', 'vtt', 'srt']:
991 params = compat_urllib_parse.urlencode({
992 'lang': original_lang,
996 'kind': caption_kind,
999 'url': caption_url + '&' + params,
1002 sub_lang_list[sub_lang] = sub_formats
1003 return sub_lang_list
1004 # An extractor error can be raise by the download process if there are
1005 # no automatic captions but there are subtitles
1006 except (KeyError, ExtractorError):
1007 self._downloader.report_warning(err_msg)
1011 def extract_id(cls, url):
1012 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1014 raise ExtractorError('Invalid URL: %s' % url)
1015 video_id = mobj.group(2)
1018 def _extract_from_m3u8(self, manifest_url, video_id):
1021 def _get_urls(_manifest):
1022 lines = _manifest.split('\n')
1023 urls = filter(lambda l: l and not l.startswith('#'),
1026 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1027 formats_urls = _get_urls(manifest)
1028 for format_url in formats_urls:
1029 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1030 url_map[itag] = format_url
1033 def _extract_annotations(self, video_id):
1034 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1035 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1037 def _parse_dash_manifest(
1038 self, video_id, dash_manifest_url, player_url, age_gate, fatal=True):
1039 def decrypt_sig(mobj):
1041 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1042 return '/signature/%s' % dec_s
1043 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
1044 dash_doc = self._download_xml(
1045 dash_manifest_url, video_id,
1046 note='Downloading DASH manifest',
1047 errnote='Could not download DASH manifest',
1050 if dash_doc is False:
1054 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
1055 mime_type = a.attrib.get('mimeType')
1056 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
1057 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
1060 if mime_type == 'text/vtt':
1061 # TODO implement WebVTT downloading
1063 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
1064 segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
1065 format_id = r.attrib['id']
1066 video_url = url_el.text
1067 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
1069 'format_id': format_id,
1071 'width': int_or_none(r.attrib.get('width')),
1072 'height': int_or_none(r.attrib.get('height')),
1073 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
1074 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
1075 'filesize': filesize,
1076 'fps': int_or_none(r.attrib.get('frameRate')),
1078 if segment_list is not None:
1080 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
1081 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
1082 'protocol': 'http_dash_segments',
1085 existing_format = next(
1086 fo for fo in formats
1087 if fo['format_id'] == format_id)
1088 except StopIteration:
1089 full_info = self._formats.get(format_id, {}).copy()
1091 codecs = r.attrib.get('codecs')
1093 if full_info.get('acodec') == 'none' and 'vcodec' not in full_info:
1094 full_info['vcodec'] = codecs
1095 elif full_info.get('vcodec') == 'none' and 'acodec' not in full_info:
1096 full_info['acodec'] = codecs
1097 formats.append(full_info)
1099 existing_format.update(f)
1101 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
1104 def _real_extract(self, url):
1105 url, smuggled_data = unsmuggle_url(url, {})
1108 'http' if self._downloader.params.get('prefer_insecure', False)
1113 parsed_url = compat_urllib_parse_urlparse(url)
1114 for component in [parsed_url.fragment, parsed_url.query]:
1115 query = compat_parse_qs(component)
1116 if start_time is None and 't' in query:
1117 start_time = parse_duration(query['t'][0])
1118 if start_time is None and 'start' in query:
1119 start_time = parse_duration(query['start'][0])
1120 if end_time is None and 'end' in query:
1121 end_time = parse_duration(query['end'][0])
1123 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1124 mobj = re.search(self._NEXT_URL_RE, url)
1126 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1127 video_id = self.extract_id(url)
1130 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1131 video_webpage = self._download_webpage(url, video_id)
1133 # Attempt to extract SWF player URL
1134 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1135 if mobj is not None:
1136 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1142 def add_dash_mpd(video_info):
1143 dash_mpd = video_info.get('dashmpd')
1144 if dash_mpd and dash_mpd[0] not in dash_mpds:
1145 dash_mpds.append(dash_mpd[0])
1148 embed_webpage = None
1150 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1152 # We simulate the access to the video from www.youtube.com/v/{video_id}
1153 # this can be viewed without login into Youtube
1154 url = proto + '://www.youtube.com/embed/%s' % video_id
1155 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1156 data = compat_urllib_parse.urlencode({
1157 'video_id': video_id,
1158 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1159 'sts': self._search_regex(
1160 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1162 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1163 video_info_webpage = self._download_webpage(
1164 video_info_url, video_id,
1165 note='Refetching age-gated info webpage',
1166 errnote='unable to download video info webpage')
1167 video_info = compat_parse_qs(video_info_webpage)
1168 add_dash_mpd(video_info)
1172 # Try looking directly into the video webpage
1173 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1175 args = ytplayer_config['args']
1176 if args.get('url_encoded_fmt_stream_map'):
1177 # Convert to the same format returned by compat_parse_qs
1178 video_info = dict((k, [v]) for k, v in args.items())
1179 add_dash_mpd(video_info)
1180 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1182 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1183 # We also try looking in get_video_info since it may contain different dashmpd
1184 # URL that points to a DASH manifest with possibly different itag set (some itags
1185 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1186 # manifest pointed by get_video_info's dashmpd).
1187 # The general idea is to take a union of itags of both DASH manifests (for example
1188 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1189 self.report_video_info_webpage_download(video_id)
1190 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1192 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1193 % (proto, video_id, el_type))
1194 video_info_webpage = self._download_webpage(
1196 video_id, note=False,
1197 errnote='unable to download video info webpage')
1198 get_video_info = compat_parse_qs(video_info_webpage)
1199 if get_video_info.get('use_cipher_signature') != ['True']:
1200 add_dash_mpd(get_video_info)
1202 video_info = get_video_info
1203 if 'token' in get_video_info:
1204 # Different get_video_info requests may report different results, e.g.
1205 # some may report video unavailability, but some may serve it without
1206 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1207 # the original webpage as well as el=info and el=embedded get_video_info
1208 # requests report video unavailability due to geo restriction while
1209 # el=detailpage succeeds and returns valid data). This is probably
1210 # due to YouTube measures against IP ranges of hosting providers.
1211 # Working around by preferring the first succeeded video_info containing
1212 # the token if no such video_info yet was found.
1213 if 'token' not in video_info:
1214 video_info = get_video_info
1216 if 'token' not in video_info:
1217 if 'reason' in video_info:
1218 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1219 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1221 raise ExtractorError('YouTube said: This video is available in %s only' % (
1222 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1224 raise ExtractorError(
1225 'YouTube said: %s' % video_info['reason'][0],
1226 expected=True, video_id=video_id)
1228 raise ExtractorError(
1229 '"token" parameter not in video info for unknown reason',
1233 if 'title' in video_info:
1234 video_title = video_info['title'][0]
1236 self._downloader.report_warning('Unable to extract video title')
1240 video_description = get_element_by_id("eow-description", video_webpage)
1241 if video_description:
1242 video_description = re.sub(r'''(?x)
1244 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1245 (?:title|href)="([^"]+)"\s+
1246 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1247 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1250 ''', r'\1', video_description)
1251 video_description = clean_html(video_description)
1253 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1255 video_description = unescapeHTML(fd_mobj.group(1))
1257 video_description = ''
1259 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1260 if not self._downloader.params.get('noplaylist'):
1263 multifeed_metadata_list = compat_urllib_parse_unquote_plus(video_info['multifeed_metadata_list'][0])
1264 for feed in multifeed_metadata_list.split(','):
1265 feed_data = compat_parse_qs(feed)
1267 '_type': 'url_transparent',
1268 'ie_key': 'Youtube',
1270 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1271 {'force_singlefeed': True}),
1272 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1274 feed_ids.append(feed_data['id'][0])
1276 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1277 % (', '.join(feed_ids), video_id))
1278 return self.playlist_result(entries, video_id, video_title, video_description)
1279 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1281 if 'view_count' in video_info:
1282 view_count = int(video_info['view_count'][0])
1286 # Check for "rental" videos
1287 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1288 raise ExtractorError('"rental" videos not supported')
1290 # Start extracting information
1291 self.report_information_extraction(video_id)
1294 if 'author' not in video_info:
1295 raise ExtractorError('Unable to extract uploader name')
1296 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1299 video_uploader_id = None
1300 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1301 if mobj is not None:
1302 video_uploader_id = mobj.group(1)
1304 self._downloader.report_warning('unable to extract uploader nickname')
1307 # We try first to get a high quality image:
1308 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1309 video_webpage, re.DOTALL)
1310 if m_thumb is not None:
1311 video_thumbnail = m_thumb.group(1)
1312 elif 'thumbnail_url' not in video_info:
1313 self._downloader.report_warning('unable to extract video thumbnail')
1314 video_thumbnail = None
1315 else: # don't panic if we can't find it
1316 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1319 upload_date = self._html_search_meta(
1320 'datePublished', video_webpage, 'upload date', default=None)
1322 upload_date = self._search_regex(
1323 [r'(?s)id="eow-date.*?>(.*?)</span>',
1324 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1325 video_webpage, 'upload date', default=None)
1327 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1328 upload_date = unified_strdate(upload_date)
1330 m_music = re.search(
1331 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1334 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1335 video_creator = clean_html(m_music.group('creator'))
1337 video_alt_title = video_creator = None
1339 m_cat_container = self._search_regex(
1340 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1341 video_webpage, 'categories', default=None)
1343 category = self._html_search_regex(
1344 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1346 video_categories = None if category is None else [category]
1348 video_categories = None
1351 unescapeHTML(m.group('content'))
1352 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1354 def _extract_count(count_name):
1355 return str_to_int(self._search_regex(
1356 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1357 % re.escape(count_name),
1358 video_webpage, count_name, default=None))
1360 like_count = _extract_count('like')
1361 dislike_count = _extract_count('dislike')
1364 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1365 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1367 if 'length_seconds' not in video_info:
1368 self._downloader.report_warning('unable to extract video duration')
1369 video_duration = None
1371 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1374 video_annotations = None
1375 if self._downloader.params.get('writeannotations', False):
1376 video_annotations = self._extract_annotations(video_id)
1378 def _map_to_format_list(urlmap):
1380 for itag, video_real_url in urlmap.items():
1383 'url': video_real_url,
1384 'player_url': player_url,
1386 if itag in self._formats:
1387 dct.update(self._formats[itag])
1391 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1392 self.report_rtmp_download()
1394 'format_id': '_rtmp',
1396 'url': video_info['conn'][0],
1397 'player_url': player_url,
1399 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1400 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1401 if 'rtmpe%3Dyes' in encoded_url_map:
1402 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1404 for url_data_str in encoded_url_map.split(','):
1405 url_data = compat_parse_qs(url_data_str)
1406 if 'itag' not in url_data or 'url' not in url_data:
1408 format_id = url_data['itag'][0]
1409 url = url_data['url'][0]
1411 if 'sig' in url_data:
1412 url += '&signature=' + url_data['sig'][0]
1413 elif 's' in url_data:
1414 encrypted_sig = url_data['s'][0]
1415 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1417 jsplayer_url_json = self._search_regex(
1419 embed_webpage if age_gate else video_webpage,
1420 'JS player URL (1)', default=None)
1421 if not jsplayer_url_json and not age_gate:
1422 # We need the embed website after all
1423 if embed_webpage is None:
1424 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1425 embed_webpage = self._download_webpage(
1426 embed_url, video_id, 'Downloading embed webpage')
1427 jsplayer_url_json = self._search_regex(
1428 ASSETS_RE, embed_webpage, 'JS player URL')
1430 player_url = json.loads(jsplayer_url_json)
1431 if player_url is None:
1432 player_url_json = self._search_regex(
1433 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1434 video_webpage, 'age gate player URL')
1435 player_url = json.loads(player_url_json)
1437 if self._downloader.params.get('verbose'):
1438 if player_url is None:
1439 player_version = 'unknown'
1440 player_desc = 'unknown'
1442 if player_url.endswith('swf'):
1443 player_version = self._search_regex(
1444 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1445 'flash player', fatal=False)
1446 player_desc = 'flash player %s' % player_version
1448 player_version = self._search_regex(
1449 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1451 'html5 player', fatal=False)
1452 player_desc = 'html5 player %s' % player_version
1454 parts_sizes = self._signature_cache_id(encrypted_sig)
1455 self.to_screen('{%s} signature length %s, %s' %
1456 (format_id, parts_sizes, player_desc))
1458 signature = self._decrypt_signature(
1459 encrypted_sig, video_id, player_url, age_gate)
1460 url += '&signature=' + signature
1461 if 'ratebypass' not in url:
1462 url += '&ratebypass=yes'
1464 # Some itags are not included in DASH manifest thus corresponding formats will
1465 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1466 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1467 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1468 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1470 'format_id': format_id,
1472 'player_url': player_url,
1473 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1474 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1477 'fps': int_or_none(url_data.get('fps', [None])[0]),
1478 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1480 type_ = url_data.get('type', [None])[0]
1482 type_split = type_.split(';')
1483 kind_ext = type_split[0].split('/')
1484 if len(kind_ext) == 2:
1485 kind, ext = kind_ext
1487 if kind in ('audio', 'video'):
1489 for mobj in re.finditer(
1490 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1491 if mobj.group('key') == 'codecs':
1492 codecs = mobj.group('val')
1495 codecs = codecs.split(',')
1496 if len(codecs) == 2:
1497 acodec, vcodec = codecs[1], codecs[0]
1499 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1504 if format_id in self._formats:
1505 dct.update(self._formats[format_id])
1507 elif video_info.get('hlsvp'):
1508 manifest_url = video_info['hlsvp'][0]
1509 url_map = self._extract_from_m3u8(manifest_url, video_id)
1510 formats = _map_to_format_list(url_map)
1511 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1512 for a_format in formats:
1513 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1515 unavailable_message = self._html_search_regex(
1516 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1517 video_webpage, 'unavailable message', default=None)
1518 if unavailable_message:
1519 raise ExtractorError(unavailable_message, expected=True)
1520 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1522 # Look for the DASH manifest
1523 if self._downloader.params.get('youtube_include_dash_manifest', True):
1524 dash_mpd_fatal = True
1525 for dash_manifest_url in dash_mpds:
1528 for df in self._parse_dash_manifest(
1529 video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
1530 # Do not overwrite DASH format found in some previous DASH manifest
1531 if df['format_id'] not in dash_formats:
1532 dash_formats[df['format_id']] = df
1533 # Additional DASH manifests may end up in HTTP Error 403 therefore
1534 # allow them to fail without bug report message if we already have
1535 # some DASH manifest succeeded. This is temporary workaround to reduce
1536 # burst of bug reports until we figure out the reason and whether it
1537 # can be fixed at all.
1538 dash_mpd_fatal = False
1539 except (ExtractorError, KeyError) as e:
1540 self.report_warning(
1541 'Skipping DASH manifest: %r' % e, video_id)
1543 # Remove the formats we found through non-DASH, they
1544 # contain less info and it can be wrong, because we use
1545 # fixed values (for example the resolution). See
1546 # https://github.com/rg3/youtube-dl/issues/5774 for an
1548 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1549 formats.extend(dash_formats.values())
1551 # Check for malformed aspect ratio
1552 stretched_m = re.search(
1553 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1556 w = float(stretched_m.group('w'))
1557 h = float(stretched_m.group('h'))
1558 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1559 # We will only process correct ratios.
1563 if f.get('vcodec') != 'none':
1564 f['stretched_ratio'] = ratio
1566 self._sort_formats(formats)
1570 'uploader': video_uploader,
1571 'uploader_id': video_uploader_id,
1572 'upload_date': upload_date,
1573 'creator': video_creator,
1574 'title': video_title,
1575 'alt_title': video_alt_title,
1576 'thumbnail': video_thumbnail,
1577 'description': video_description,
1578 'categories': video_categories,
1580 'subtitles': video_subtitles,
1581 'automatic_captions': automatic_captions,
1582 'duration': video_duration,
1583 'age_limit': 18 if age_gate else 0,
1584 'annotations': video_annotations,
1585 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1586 'view_count': view_count,
1587 'like_count': like_count,
1588 'dislike_count': dislike_count,
1589 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1592 'start_time': start_time,
1593 'end_time': end_time,
1597 class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor):
1598 IE_DESC = 'YouTube.com playlists'
1599 _VALID_URL = r"""(?x)(?:
1604 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1605 \? (?:.*?[&;])*? (?:p|a|list)=
1609 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1610 # Top tracks, they can also include dots
1615 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1617 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1618 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1619 IE_NAME = 'youtube:playlist'
1621 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1623 'title': 'ytdl test PL',
1624 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1626 'playlist_count': 3,
1628 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1630 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1631 'title': 'YDL_Empty_List',
1633 'playlist_count': 0,
1635 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1636 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1638 'title': '29C3: Not my department',
1639 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1641 'playlist_count': 95,
1643 'note': 'issue #673',
1644 'url': 'PLBB231211A4F62143',
1646 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1647 'id': 'PLBB231211A4F62143',
1649 'playlist_mincount': 26,
1651 'note': 'Large playlist',
1652 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1654 'title': 'Uploads from Cauchemar',
1655 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1657 'playlist_mincount': 799,
1659 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1661 'title': 'YDL_safe_search',
1662 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1664 'playlist_count': 2,
1667 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1668 'playlist_count': 4,
1671 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1674 'note': 'Embedded SWF player',
1675 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1676 'playlist_count': 4,
1679 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1682 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1683 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1685 'title': 'Uploads from Interstellar Movie',
1686 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1688 'playlist_mincout': 21,
1691 def _real_initialize(self):
1694 def _extract_mix(self, playlist_id):
1695 # The mixes are generated from a single video
1696 # the id of the playlist is just 'RD' + video_id
1697 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1698 webpage = self._download_webpage(
1699 url, playlist_id, 'Downloading Youtube mix')
1700 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1702 search_title('playlist-title') or
1703 search_title('title long-title') or
1704 search_title('title'))
1705 title = clean_html(title_span)
1706 ids = orderedSet(re.findall(
1707 r'''(?xs)data-video-username=".*?".*?
1708 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1710 url_results = self._ids_to_results(ids)
1712 return self.playlist_result(url_results, playlist_id, title)
1714 def _extract_playlist(self, playlist_id):
1715 url = self._TEMPLATE_URL % playlist_id
1716 page = self._download_webpage(url, playlist_id)
1718 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1719 match = match.strip()
1720 # Check if the playlist exists or is private
1721 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1722 raise ExtractorError(
1723 'The playlist doesn\'t exist or is private, use --username or '
1724 '--netrc to access it.',
1726 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1727 raise ExtractorError(
1728 'Invalid parameters. Maybe URL is incorrect.',
1730 elif re.match(r'[^<]*Choose your language[^<]*', match):
1733 self.report_warning('Youtube gives an alert message: ' + match)
1735 playlist_title = self._html_search_regex(
1736 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1739 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1741 def _real_extract(self, url):
1742 # Extract playlist id
1743 mobj = re.match(self._VALID_URL, url)
1745 raise ExtractorError('Invalid URL: %s' % url)
1746 playlist_id = mobj.group(1) or mobj.group(2)
1748 # Check if it's a video-specific URL
1749 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1750 if 'v' in query_dict:
1751 video_id = query_dict['v'][0]
1752 if self._downloader.params.get('noplaylist'):
1753 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1754 return self.url_result(video_id, 'Youtube', video_id=video_id)
1756 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1758 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1759 # Mixes require a custom extraction process
1760 return self._extract_mix(playlist_id)
1762 return self._extract_playlist(playlist_id)
1765 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1766 IE_DESC = 'YouTube.com channels'
1767 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1768 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1769 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1770 IE_NAME = 'youtube:channel'
1772 'note': 'paginated channel',
1773 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1774 'playlist_mincount': 91,
1776 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1777 'title': 'Uploads from lex will',
1780 'note': 'Age restricted channel',
1781 # from https://www.youtube.com/user/DeusExOfficial
1782 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1783 'playlist_mincount': 64,
1785 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1786 'title': 'Uploads from Deus Ex',
1791 def suitable(cls, url):
1792 return False if YoutubePlaylistsIE.suitable(url) else super(YoutubeChannelIE, cls).suitable(url)
1794 def _real_extract(self, url):
1795 channel_id = self._match_id(url)
1797 url = self._TEMPLATE_URL % channel_id
1799 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1800 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1801 # otherwise fallback on channel by page extraction
1802 channel_page = self._download_webpage(
1803 url + '?view=57', channel_id,
1804 'Downloading channel page', fatal=False)
1805 if channel_page is False:
1806 channel_playlist_id = False
1808 channel_playlist_id = self._html_search_meta(
1809 'channelId', channel_page, 'channel id', default=None)
1810 if not channel_playlist_id:
1811 channel_playlist_id = self._search_regex(
1812 r'data-(?:channel-external-|yt)id="([^"]+)"',
1813 channel_page, 'channel id', default=None)
1814 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1815 playlist_id = 'UU' + channel_playlist_id[2:]
1816 return self.url_result(
1817 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1819 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1820 autogenerated = re.search(r'''(?x)
1822 channel-header-autogenerated-label|
1823 yt-channel-title-autogenerated
1824 )[^"]*"''', channel_page) is not None
1827 # The videos are contained in a single page
1828 # the ajax pages can't be used, they are empty
1831 video_id, 'Youtube', video_id=video_id,
1832 video_title=video_title)
1833 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1834 return self.playlist_result(entries, channel_id)
1836 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1839 class YoutubeUserIE(YoutubeChannelIE):
1840 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1841 _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_-]+)'
1842 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1843 IE_NAME = 'youtube:user'
1846 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1847 'playlist_mincount': 320,
1849 'title': 'TheLinuxFoundation',
1852 'url': 'ytuser:phihag',
1853 'only_matching': True,
1857 def suitable(cls, url):
1858 # Don't return True if the url can be extracted with other youtube
1859 # extractor, the regex would is too permissive and it would match.
1860 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1861 if any(ie.suitable(url) for ie in other_ies):
1864 return super(YoutubeUserIE, cls).suitable(url)
1867 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1868 IE_DESC = 'YouTube.com user/channel playlists'
1869 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
1870 IE_NAME = 'youtube:playlists'
1873 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1874 'playlist_mincount': 4,
1876 'id': 'ThirstForScience',
1877 'title': 'Thirst for Science',
1880 # with "Load more" button
1881 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1882 'playlist_mincount': 70,
1885 'title': 'Игорь Клейнер',
1888 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
1889 'playlist_mincount': 17,
1891 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
1892 'title': 'Chem Player',
1897 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1898 IE_DESC = 'YouTube.com searches'
1899 # there doesn't appear to be a real limit, for example if you search for
1900 # 'python' you get more than 8.000.000 results
1901 _MAX_RESULTS = float('inf')
1902 IE_NAME = 'youtube:search'
1903 _SEARCH_KEY = 'ytsearch'
1904 _EXTRA_QUERY_ARGS = {}
1907 def _get_n_results(self, query, n):
1908 """Get a specified number of results for a query"""
1913 for pagenum in itertools.count(1):
1915 'search_query': query.encode('utf-8'),
1919 url_query.update(self._EXTRA_QUERY_ARGS)
1920 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1921 data = self._download_json(
1922 result_url, video_id='query "%s"' % query,
1923 note='Downloading page %s' % pagenum,
1924 errnote='Unable to download API page')
1925 html_content = data[1]['body']['content']
1927 if 'class="search-message' in html_content:
1928 raise ExtractorError(
1929 '[youtube] No video results', expected=True)
1931 new_videos = self._ids_to_results(orderedSet(re.findall(
1932 r'href="/watch\?v=(.{11})', html_content)))
1933 videos += new_videos
1934 if not new_videos or len(videos) > limit:
1939 return self.playlist_result(videos, query)
1942 class YoutubeSearchDateIE(YoutubeSearchIE):
1943 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1944 _SEARCH_KEY = 'ytsearchdate'
1945 IE_DESC = 'YouTube.com searches, newest videos first'
1946 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1949 class YoutubeSearchURLIE(InfoExtractor):
1950 IE_DESC = 'YouTube.com search URLs'
1951 IE_NAME = 'youtube:search_url'
1952 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1954 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1955 'playlist_mincount': 5,
1957 'title': 'youtube-dl test video',
1961 def _real_extract(self, url):
1962 mobj = re.match(self._VALID_URL, url)
1963 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
1965 webpage = self._download_webpage(url, query)
1966 result_code = self._search_regex(
1967 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1969 part_codes = re.findall(
1970 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
1972 for part_code in part_codes:
1973 part_title = self._html_search_regex(
1974 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1975 part_url_snippet = self._html_search_regex(
1976 r'(?s)href="([^"]+)"', part_code, 'item URL')
1977 part_url = compat_urlparse.urljoin(
1978 'https://www.youtube.com/', part_url_snippet)
1982 'title': part_title,
1986 '_type': 'playlist',
1992 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
1993 IE_DESC = 'YouTube.com (multi-season) shows'
1994 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1995 IE_NAME = 'youtube:show'
1997 'url': 'https://www.youtube.com/show/airdisasters',
1998 'playlist_mincount': 5,
2000 'id': 'airdisasters',
2001 'title': 'Air Disasters',
2005 def _real_extract(self, url):
2006 playlist_id = self._match_id(url)
2007 return super(YoutubeShowIE, self)._real_extract(
2008 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2011 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2013 Base class for feed extractors
2014 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2016 _LOGIN_REQUIRED = True
2020 return 'youtube:%s' % self._FEED_NAME
2022 def _real_initialize(self):
2025 def _real_extract(self, url):
2026 page = self._download_webpage(
2027 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2029 # The extraction process is the same as for playlists, but the regex
2030 # for the video ids doesn't contain an index
2032 more_widget_html = content_html = page
2033 for page_num in itertools.count(1):
2034 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2036 # 'recommended' feed has infinite 'load more' and each new portion spins
2037 # the same videos in (sometimes) slightly different order, so we'll check
2038 # for unicity and break when portion has no new videos
2039 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2045 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2049 more = self._download_json(
2050 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2051 'Downloading page #%s' % page_num,
2052 transform_source=uppercase_escape)
2053 content_html = more['content_html']
2054 more_widget_html = more['load_more_widget_html']
2056 return self.playlist_result(
2057 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2060 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2061 IE_NAME = 'youtube:watchlater'
2062 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2063 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
2065 _TESTS = [] # override PlaylistIE tests
2067 def _real_extract(self, url):
2068 return self._extract_playlist('WL')
2071 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2072 IE_NAME = 'youtube:favorites'
2073 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2074 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2075 _LOGIN_REQUIRED = True
2077 def _real_extract(self, url):
2078 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2079 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2080 return self.url_result(playlist_id, 'YoutubePlaylist')
2083 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2084 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2085 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2086 _FEED_NAME = 'recommended'
2087 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2090 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2091 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2092 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2093 _FEED_NAME = 'subscriptions'
2094 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2097 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2098 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2099 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2100 _FEED_NAME = 'history'
2101 _PLAYLIST_TITLE = 'Youtube History'
2104 class YoutubeTruncatedURLIE(InfoExtractor):
2105 IE_NAME = 'youtube:truncated_url'
2106 IE_DESC = False # Do not list
2107 _VALID_URL = r'''(?x)
2109 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2112 annotation_id=annotation_[^&]+|
2118 attribution_link\?a=[^&]+
2124 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2125 'only_matching': True,
2127 'url': 'http://www.youtube.com/watch?',
2128 'only_matching': True,
2130 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2131 'only_matching': True,
2133 'url': 'https://www.youtube.com/watch?feature=foo',
2134 'only_matching': True,
2136 'url': 'https://www.youtube.com/watch?hl=en-GB',
2137 'only_matching': True,
2139 'url': 'https://www.youtube.com/watch?t=2372',
2140 'only_matching': True,
2143 def _real_extract(self, url):
2144 raise ExtractorError(
2145 'Did you forget to quote the URL? Remember that & is a meta '
2146 'character in most shells, so you want to put the URL in quotes, '
2148 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2149 ' or simply youtube-dl BaW_jenozKc .',
2153 class YoutubeTruncatedIDIE(InfoExtractor):
2154 IE_NAME = 'youtube:truncated_id'
2155 IE_DESC = False # Do not list
2156 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2159 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2160 'only_matching': True,
2163 def _real_extract(self, url):
2164 video_id = self._match_id(url)
2165 raise ExtractorError(
2166 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),