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,
23 compat_urllib_request,
32 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 = compat_urllib_request.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 = compat_urllib_request.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)
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 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
350 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
351 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
352 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
353 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'vp9'},
354 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'},
357 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
358 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
360 # Dash webm audio with opus inside
361 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
362 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
363 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
366 '_rtmp': {'protocol': 'rtmp'},
372 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
376 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
377 'uploader': 'Philipp Hagemeister',
378 'uploader_id': 'phihag',
379 'upload_date': '20121002',
380 '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 .',
381 'categories': ['Science & Technology'],
382 'tags': ['youtube-dl'],
384 'dislike_count': int,
390 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
391 'note': 'Test generic use_cipher_signature video (#897)',
395 'upload_date': '20120506',
396 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
397 'description': 'md5:782e8651347686cba06e58f71ab51773',
398 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
399 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
400 'iconic ep', 'iconic', 'love', 'it'],
401 'uploader': 'Icona Pop',
402 'uploader_id': 'IconaPop',
406 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
407 'note': 'Test VEVO video with age protection (#956)',
411 'upload_date': '20130703',
412 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
413 'description': 'md5:64249768eec3bc4276236606ea996373',
414 'uploader': 'justintimberlakeVEVO',
415 'uploader_id': 'justintimberlakeVEVO',
420 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
421 'note': 'Embed-only video (#1746)',
425 'upload_date': '20120608',
426 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
427 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
428 'uploader': 'SET India',
429 'uploader_id': 'setindia'
433 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY',
434 'note': 'Use the first video ID in the URL',
438 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
439 'uploader': 'Philipp Hagemeister',
440 'uploader_id': 'phihag',
441 'upload_date': '20121002',
442 '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 .',
443 'categories': ['Science & Technology'],
444 'tags': ['youtube-dl'],
446 'dislike_count': int,
449 'skip_download': True,
453 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
454 'note': '256k DASH audio (format 141) via DASH manifest',
458 'upload_date': '20121002',
459 'uploader_id': '8KVIDEO',
461 'uploader': '8KVIDEO',
462 'title': 'UHDTV TEST 8K VIDEO.mp4'
465 'youtube_include_dash_manifest': True,
469 # DASH manifest with encrypted signature
471 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
475 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
476 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
477 'uploader': 'AfrojackVEVO',
478 'uploader_id': 'AfrojackVEVO',
479 'upload_date': '20131011',
482 'youtube_include_dash_manifest': True,
486 # JS player signature function name containing $
488 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
492 'title': 'Taylor Swift - Shake It Off',
493 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
494 'uploader': 'TaylorSwiftVEVO',
495 'uploader_id': 'TaylorSwiftVEVO',
496 'upload_date': '20140818',
499 'youtube_include_dash_manifest': True,
505 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
509 'upload_date': '20100909',
510 'uploader': 'The Amazing Atheist',
511 'uploader_id': 'TheAmazingAtheist',
512 'title': 'Burning Everyone\'s Koran',
513 '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',
516 # Normal age-gate video (No vevo, embed allowed)
518 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
522 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
523 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
524 'uploader': 'The Witcher',
525 'uploader_id': 'WitcherGame',
526 'upload_date': '20140605',
530 # Age-gate video with encrypted signature
532 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
536 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
537 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
538 'uploader': 'LloydVEVO',
539 'uploader_id': 'LloydVEVO',
540 'upload_date': '20110629',
544 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
546 'url': '__2ABJjxzNo',
550 'upload_date': '20100430',
551 'uploader_id': 'deadmau5',
552 'description': 'md5:12c56784b8032162bb936a5f76d55360',
553 'uploader': 'deadmau5',
554 'title': 'Deadmau5 - Some Chords (HD)',
556 'expected_warnings': [
557 'DASH manifest missing',
560 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
562 'url': 'lqQg6PlCWgI',
566 'upload_date': '20120724',
567 'uploader_id': 'olympic',
568 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
569 'uploader': 'Olympics',
570 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
573 'skip_download': 'requires avconv',
578 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
582 'stretched_ratio': 16 / 9.,
583 'upload_date': '20110310',
584 'uploader_id': 'AllenMeow',
585 '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',
587 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
590 # url_encoded_fmt_stream_map is empty string
592 'url': 'qEJwOuvDf7I',
596 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
598 'upload_date': '20150404',
599 'uploader_id': 'spbelect',
600 'uploader': 'Наблюдатели Петербурга',
603 'skip_download': 'requires avconv',
606 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
608 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
612 'title': 'md5:7b81415841e02ecd4313668cde88737a',
613 'description': 'md5:116377fd2963b81ec4ce64b542173306',
614 'upload_date': '20150625',
615 'uploader_id': 'dorappi2000',
616 'uploader': 'dorappi2000',
617 'formats': 'mincount:33',
620 # DASH manifest with segment_list
622 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
623 'md5': '8ce563a1d667b599d21064e982ab9e31',
627 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
628 'uploader': 'Airtek',
629 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
630 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
631 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
634 'youtube_include_dash_manifest': True,
635 'format': '135', # bestvideo
639 # Multifeed videos (multiple cameras), URL is for Main Camera
640 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
643 'title': 'teamPGP: Rocket League Noob Stream',
644 'description': 'md5:dc7872fb300e143831327f1bae3af010',
650 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
651 'description': 'md5:dc7872fb300e143831327f1bae3af010',
652 'upload_date': '20150721',
653 'uploader': 'Beer Games Beer',
654 'uploader_id': 'beergamesbeer',
660 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
661 'description': 'md5:dc7872fb300e143831327f1bae3af010',
662 'upload_date': '20150721',
663 'uploader': 'Beer Games Beer',
664 'uploader_id': 'beergamesbeer',
670 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
671 'description': 'md5:dc7872fb300e143831327f1bae3af010',
672 'upload_date': '20150721',
673 'uploader': 'Beer Games Beer',
674 'uploader_id': 'beergamesbeer',
680 'title': 'teamPGP: Rocket League Noob Stream (zim)',
681 'description': 'md5:dc7872fb300e143831327f1bae3af010',
682 'upload_date': '20150721',
683 'uploader': 'Beer Games Beer',
684 'uploader_id': 'beergamesbeer',
688 'skip_download': True,
692 'url': 'http://vid.plus/FlRa-iH7PGw',
693 'only_matching': True,
696 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
697 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
701 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
702 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
703 'upload_date': '20151119',
704 'uploader_id': 'IronSoulElf',
705 'uploader': 'IronSoulElf',
708 'skip_download': True,
712 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
713 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
714 'only_matching': True,
718 def __init__(self, *args, **kwargs):
719 super(YoutubeIE, self).__init__(*args, **kwargs)
720 self._player_cache = {}
722 def report_video_info_webpage_download(self, video_id):
723 """Report attempt to download video info webpage."""
724 self.to_screen('%s: Downloading video info webpage' % video_id)
726 def report_information_extraction(self, video_id):
727 """Report attempt to extract video information."""
728 self.to_screen('%s: Extracting video information' % video_id)
730 def report_unavailable_format(self, video_id, format):
731 """Report extracted video URL."""
732 self.to_screen('%s: Format %s not available' % (video_id, format))
734 def report_rtmp_download(self):
735 """Indicate the download will use the RTMP protocol."""
736 self.to_screen('RTMP download detected')
738 def _signature_cache_id(self, example_sig):
739 """ Return a string representation of a signature """
740 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
742 def _extract_signature_function(self, video_id, player_url, example_sig):
744 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
747 raise ExtractorError('Cannot identify player %r' % player_url)
748 player_type = id_m.group('ext')
749 player_id = id_m.group('id')
751 # Read from filesystem cache
752 func_id = '%s_%s_%s' % (
753 player_type, player_id, self._signature_cache_id(example_sig))
754 assert os.path.basename(func_id) == func_id
756 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
757 if cache_spec is not None:
758 return lambda s: ''.join(s[i] for i in cache_spec)
761 'Downloading player %s' % player_url
762 if self._downloader.params.get('verbose') else
763 'Downloading %s player %s' % (player_type, player_id)
765 if player_type == 'js':
766 code = self._download_webpage(
767 player_url, video_id,
769 errnote='Download of %s failed' % player_url)
770 res = self._parse_sig_js(code)
771 elif player_type == 'swf':
772 urlh = self._request_webpage(
773 player_url, video_id,
775 errnote='Download of %s failed' % player_url)
777 res = self._parse_sig_swf(code)
779 assert False, 'Invalid player type %r' % player_type
781 test_string = ''.join(map(compat_chr, range(len(example_sig))))
782 cache_res = res(test_string)
783 cache_spec = [ord(c) for c in cache_res]
785 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
788 def _print_sig_code(self, func, example_sig):
789 def gen_sig_code(idxs):
790 def _genslice(start, end, step):
791 starts = '' if start == 0 else str(start)
792 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
793 steps = '' if step == 1 else (':%d' % step)
794 return 's[%s%s%s]' % (starts, ends, steps)
797 # Quelch pyflakes warnings - start will be set when step is set
798 start = '(Never used)'
799 for i, prev in zip(idxs[1:], idxs[:-1]):
803 yield _genslice(start, prev, step)
806 if i - prev in [-1, 1]:
815 yield _genslice(start, i, step)
817 test_string = ''.join(map(compat_chr, range(len(example_sig))))
818 cache_res = func(test_string)
819 cache_spec = [ord(c) for c in cache_res]
820 expr_code = ' + '.join(gen_sig_code(cache_spec))
821 signature_id_tuple = '(%s)' % (
822 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
823 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
824 ' return %s\n') % (signature_id_tuple, expr_code)
825 self.to_screen('Extracted signature function:\n' + code)
827 def _parse_sig_js(self, jscode):
828 funcname = self._search_regex(
829 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
830 'Initial JS player signature function name')
832 jsi = JSInterpreter(jscode)
833 initial_function = jsi.extract_function(funcname)
834 return lambda s: initial_function([s])
836 def _parse_sig_swf(self, file_contents):
837 swfi = SWFInterpreter(file_contents)
838 TARGET_CLASSNAME = 'SignatureDecipher'
839 searched_class = swfi.extract_class(TARGET_CLASSNAME)
840 initial_function = swfi.extract_function(searched_class, 'decipher')
841 return lambda s: initial_function([s])
843 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
844 """Turn the encrypted s field into a working signature"""
846 if player_url is None:
847 raise ExtractorError('Cannot decrypt signature without player_url')
849 if player_url.startswith('//'):
850 player_url = 'https:' + player_url
852 player_id = (player_url, self._signature_cache_id(s))
853 if player_id not in self._player_cache:
854 func = self._extract_signature_function(
855 video_id, player_url, s
857 self._player_cache[player_id] = func
858 func = self._player_cache[player_id]
859 if self._downloader.params.get('youtube_print_sig_code'):
860 self._print_sig_code(func, s)
862 except Exception as e:
863 tb = traceback.format_exc()
864 raise ExtractorError(
865 'Signature extraction failed: ' + tb, cause=e)
867 def _get_subtitles(self, video_id, webpage):
869 subs_doc = self._download_xml(
870 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
871 video_id, note=False)
872 except ExtractorError as err:
873 self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
877 for track in subs_doc.findall('track'):
878 lang = track.attrib['lang_code']
879 if lang in sub_lang_list:
882 for ext in ['sbv', 'vtt', 'srt']:
883 params = compat_urllib_parse.urlencode({
887 'name': track.attrib['name'].encode('utf-8'),
890 'url': 'https://www.youtube.com/api/timedtext?' + params,
893 sub_lang_list[lang] = sub_formats
894 if not sub_lang_list:
895 self._downloader.report_warning('video doesn\'t have subtitles')
899 def _get_ytplayer_config(self, video_id, webpage):
901 # User data may contain arbitrary character sequences that may affect
902 # JSON extraction with regex, e.g. when '};' is contained the second
903 # regex won't capture the whole JSON. Yet working around by trying more
904 # concrete regex first keeping in mind proper quoted string handling
905 # to be implemented in future that will replace this workaround (see
906 # https://github.com/rg3/youtube-dl/issues/7468,
907 # https://github.com/rg3/youtube-dl/pull/7599)
908 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
909 r';ytplayer\.config\s*=\s*({.+?});',
911 config = self._search_regex(
912 patterns, webpage, 'ytplayer.config', default=None)
914 return self._parse_json(
915 uppercase_escape(config), video_id, fatal=False)
917 def _get_automatic_captions(self, video_id, webpage):
918 """We need the webpage for getting the captions url, pass it as an
919 argument to speed up the process."""
920 self.to_screen('%s: Looking for automatic captions' % video_id)
921 player_config = self._get_ytplayer_config(video_id, webpage)
922 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
923 if not player_config:
924 self._downloader.report_warning(err_msg)
927 args = player_config['args']
928 caption_url = args['ttsurl']
929 timestamp = args['timestamp']
930 # We get the available subtitles
931 list_params = compat_urllib_parse.urlencode({
936 list_url = caption_url + '&' + list_params
937 caption_list = self._download_xml(list_url, video_id)
938 original_lang_node = caption_list.find('track')
939 if original_lang_node is None:
940 self._downloader.report_warning('Video doesn\'t have automatic captions')
942 original_lang = original_lang_node.attrib['lang_code']
943 caption_kind = original_lang_node.attrib.get('kind', '')
946 for lang_node in caption_list.findall('target'):
947 sub_lang = lang_node.attrib['lang_code']
949 for ext in ['sbv', 'vtt', 'srt']:
950 params = compat_urllib_parse.urlencode({
951 'lang': original_lang,
955 'kind': caption_kind,
958 'url': caption_url + '&' + params,
961 sub_lang_list[sub_lang] = sub_formats
963 # An extractor error can be raise by the download process if there are
964 # no automatic captions but there are subtitles
965 except (KeyError, ExtractorError):
966 self._downloader.report_warning(err_msg)
970 def extract_id(cls, url):
971 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
973 raise ExtractorError('Invalid URL: %s' % url)
974 video_id = mobj.group(2)
977 def _extract_from_m3u8(self, manifest_url, video_id):
980 def _get_urls(_manifest):
981 lines = _manifest.split('\n')
982 urls = filter(lambda l: l and not l.startswith('#'),
985 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
986 formats_urls = _get_urls(manifest)
987 for format_url in formats_urls:
988 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
989 url_map[itag] = format_url
992 def _extract_annotations(self, video_id):
993 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
994 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
996 def _parse_dash_manifest(
997 self, video_id, dash_manifest_url, player_url, age_gate, fatal=True):
998 def decrypt_sig(mobj):
1000 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1001 return '/signature/%s' % dec_s
1002 dash_manifest_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, dash_manifest_url)
1003 dash_doc = self._download_xml(
1004 dash_manifest_url, video_id,
1005 note='Downloading DASH manifest',
1006 errnote='Could not download DASH manifest',
1009 if dash_doc is False:
1013 for a in dash_doc.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
1014 mime_type = a.attrib.get('mimeType')
1015 for r in a.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
1016 url_el = r.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
1019 if mime_type == 'text/vtt':
1020 # TODO implement WebVTT downloading
1022 elif mime_type.startswith('audio/') or mime_type.startswith('video/'):
1023 segment_list = r.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
1024 format_id = r.attrib['id']
1025 video_url = url_el.text
1026 filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength'))
1028 'format_id': format_id,
1030 'width': int_or_none(r.attrib.get('width')),
1031 'height': int_or_none(r.attrib.get('height')),
1032 'tbr': int_or_none(r.attrib.get('bandwidth'), 1000),
1033 'asr': int_or_none(r.attrib.get('audioSamplingRate')),
1034 'filesize': filesize,
1035 'fps': int_or_none(r.attrib.get('frameRate')),
1037 if segment_list is not None:
1039 'initialization_url': segment_list.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib['sourceURL'],
1040 'segment_urls': [segment.attrib.get('media') for segment in segment_list.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
1041 'protocol': 'http_dash_segments',
1044 existing_format = next(
1045 fo for fo in formats
1046 if fo['format_id'] == format_id)
1047 except StopIteration:
1048 full_info = self._formats.get(format_id, {}).copy()
1050 codecs = r.attrib.get('codecs')
1052 if full_info.get('acodec') == 'none' and 'vcodec' not in full_info:
1053 full_info['vcodec'] = codecs
1054 elif full_info.get('vcodec') == 'none' and 'acodec' not in full_info:
1055 full_info['acodec'] = codecs
1056 formats.append(full_info)
1058 existing_format.update(f)
1060 self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
1063 def _real_extract(self, url):
1064 url, smuggled_data = unsmuggle_url(url, {})
1067 'http' if self._downloader.params.get('prefer_insecure', False)
1072 parsed_url = compat_urllib_parse_urlparse(url)
1073 for component in [parsed_url.fragment, parsed_url.query]:
1074 query = compat_parse_qs(component)
1075 if start_time is None and 't' in query:
1076 start_time = parse_duration(query['t'][0])
1077 if start_time is None and 'start' in query:
1078 start_time = parse_duration(query['start'][0])
1079 if end_time is None and 'end' in query:
1080 end_time = parse_duration(query['end'][0])
1082 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1083 mobj = re.search(self._NEXT_URL_RE, url)
1085 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1086 video_id = self.extract_id(url)
1089 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1090 video_webpage = self._download_webpage(url, video_id)
1092 # Attempt to extract SWF player URL
1093 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1094 if mobj is not None:
1095 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1101 def add_dash_mpd(video_info):
1102 dash_mpd = video_info.get('dashmpd')
1103 if dash_mpd and dash_mpd[0] not in dash_mpds:
1104 dash_mpds.append(dash_mpd[0])
1107 embed_webpage = None
1109 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1111 # We simulate the access to the video from www.youtube.com/v/{video_id}
1112 # this can be viewed without login into Youtube
1113 url = proto + '://www.youtube.com/embed/%s' % video_id
1114 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1115 data = compat_urllib_parse.urlencode({
1116 'video_id': video_id,
1117 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1118 'sts': self._search_regex(
1119 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1121 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1122 video_info_webpage = self._download_webpage(
1123 video_info_url, video_id,
1124 note='Refetching age-gated info webpage',
1125 errnote='unable to download video info webpage')
1126 video_info = compat_parse_qs(video_info_webpage)
1127 add_dash_mpd(video_info)
1131 # Try looking directly into the video webpage
1132 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1134 args = ytplayer_config['args']
1135 if args.get('url_encoded_fmt_stream_map'):
1136 # Convert to the same format returned by compat_parse_qs
1137 video_info = dict((k, [v]) for k, v in args.items())
1138 add_dash_mpd(video_info)
1139 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1141 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1142 # We also try looking in get_video_info since it may contain different dashmpd
1143 # URL that points to a DASH manifest with possibly different itag set (some itags
1144 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1145 # manifest pointed by get_video_info's dashmpd).
1146 # The general idea is to take a union of itags of both DASH manifests (for example
1147 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1148 self.report_video_info_webpage_download(video_id)
1149 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1151 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1152 % (proto, video_id, el_type))
1153 video_info_webpage = self._download_webpage(
1155 video_id, note=False,
1156 errnote='unable to download video info webpage')
1157 get_video_info = compat_parse_qs(video_info_webpage)
1158 if get_video_info.get('use_cipher_signature') != ['True']:
1159 add_dash_mpd(get_video_info)
1161 video_info = get_video_info
1162 if 'token' in get_video_info:
1163 # Different get_video_info requests may report different results, e.g.
1164 # some may report video unavailability, but some may serve it without
1165 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1166 # the original webpage as well as el=info and el=embedded get_video_info
1167 # requests report video unavailability due to geo restriction while
1168 # el=detailpage succeeds and returns valid data). This is probably
1169 # due to YouTube measures against IP ranges of hosting providers.
1170 # Working around by preferring the first succeeded video_info containing
1171 # the token if no such video_info yet was found.
1172 if 'token' not in video_info:
1173 video_info = get_video_info
1175 if 'token' not in video_info:
1176 if 'reason' in video_info:
1177 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1178 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1180 raise ExtractorError('YouTube said: This video is available in %s only' % (
1181 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1183 raise ExtractorError(
1184 'YouTube said: %s' % video_info['reason'][0],
1185 expected=True, video_id=video_id)
1187 raise ExtractorError(
1188 '"token" parameter not in video info for unknown reason',
1192 if 'title' in video_info:
1193 video_title = video_info['title'][0]
1195 self._downloader.report_warning('Unable to extract video title')
1199 video_description = get_element_by_id("eow-description", video_webpage)
1200 if video_description:
1201 video_description = re.sub(r'''(?x)
1203 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1205 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1206 class="yt-uix-redirect-link"\s*>
1209 ''', r'\1', video_description)
1210 video_description = clean_html(video_description)
1212 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1214 video_description = unescapeHTML(fd_mobj.group(1))
1216 video_description = ''
1218 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1219 if not self._downloader.params.get('noplaylist'):
1222 multifeed_metadata_list = compat_urllib_parse_unquote_plus(video_info['multifeed_metadata_list'][0])
1223 for feed in multifeed_metadata_list.split(','):
1224 feed_data = compat_parse_qs(feed)
1226 '_type': 'url_transparent',
1227 'ie_key': 'Youtube',
1229 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1230 {'force_singlefeed': True}),
1231 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1233 feed_ids.append(feed_data['id'][0])
1235 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1236 % (', '.join(feed_ids), video_id))
1237 return self.playlist_result(entries, video_id, video_title, video_description)
1238 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1240 if 'view_count' in video_info:
1241 view_count = int(video_info['view_count'][0])
1245 # Check for "rental" videos
1246 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1247 raise ExtractorError('"rental" videos not supported')
1249 # Start extracting information
1250 self.report_information_extraction(video_id)
1253 if 'author' not in video_info:
1254 raise ExtractorError('Unable to extract uploader name')
1255 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1258 video_uploader_id = None
1259 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1260 if mobj is not None:
1261 video_uploader_id = mobj.group(1)
1263 self._downloader.report_warning('unable to extract uploader nickname')
1266 # We try first to get a high quality image:
1267 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1268 video_webpage, re.DOTALL)
1269 if m_thumb is not None:
1270 video_thumbnail = m_thumb.group(1)
1271 elif 'thumbnail_url' not in video_info:
1272 self._downloader.report_warning('unable to extract video thumbnail')
1273 video_thumbnail = None
1274 else: # don't panic if we can't find it
1275 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1278 upload_date = self._html_search_meta(
1279 'datePublished', video_webpage, 'upload date', default=None)
1281 upload_date = self._search_regex(
1282 [r'(?s)id="eow-date.*?>(.*?)</span>',
1283 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1284 video_webpage, 'upload date', default=None)
1286 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1287 upload_date = unified_strdate(upload_date)
1289 m_cat_container = self._search_regex(
1290 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1291 video_webpage, 'categories', default=None)
1293 category = self._html_search_regex(
1294 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1296 video_categories = None if category is None else [category]
1298 video_categories = None
1301 unescapeHTML(m.group('content'))
1302 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1304 def _extract_count(count_name):
1305 return str_to_int(self._search_regex(
1306 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1307 % re.escape(count_name),
1308 video_webpage, count_name, default=None))
1310 like_count = _extract_count('like')
1311 dislike_count = _extract_count('dislike')
1314 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1315 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1317 if 'length_seconds' not in video_info:
1318 self._downloader.report_warning('unable to extract video duration')
1319 video_duration = None
1321 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1324 video_annotations = None
1325 if self._downloader.params.get('writeannotations', False):
1326 video_annotations = self._extract_annotations(video_id)
1328 def _map_to_format_list(urlmap):
1330 for itag, video_real_url in urlmap.items():
1333 'url': video_real_url,
1334 'player_url': player_url,
1336 if itag in self._formats:
1337 dct.update(self._formats[itag])
1341 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1342 self.report_rtmp_download()
1344 'format_id': '_rtmp',
1346 'url': video_info['conn'][0],
1347 'player_url': player_url,
1349 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1350 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1351 if 'rtmpe%3Dyes' in encoded_url_map:
1352 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1354 for url_data_str in encoded_url_map.split(','):
1355 url_data = compat_parse_qs(url_data_str)
1356 if 'itag' not in url_data or 'url' not in url_data:
1358 format_id = url_data['itag'][0]
1359 url = url_data['url'][0]
1361 if 'sig' in url_data:
1362 url += '&signature=' + url_data['sig'][0]
1363 elif 's' in url_data:
1364 encrypted_sig = url_data['s'][0]
1365 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1367 jsplayer_url_json = self._search_regex(
1369 embed_webpage if age_gate else video_webpage,
1370 'JS player URL (1)', default=None)
1371 if not jsplayer_url_json and not age_gate:
1372 # We need the embed website after all
1373 if embed_webpage is None:
1374 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1375 embed_webpage = self._download_webpage(
1376 embed_url, video_id, 'Downloading embed webpage')
1377 jsplayer_url_json = self._search_regex(
1378 ASSETS_RE, embed_webpage, 'JS player URL')
1380 player_url = json.loads(jsplayer_url_json)
1381 if player_url is None:
1382 player_url_json = self._search_regex(
1383 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1384 video_webpage, 'age gate player URL')
1385 player_url = json.loads(player_url_json)
1387 if self._downloader.params.get('verbose'):
1388 if player_url is None:
1389 player_version = 'unknown'
1390 player_desc = 'unknown'
1392 if player_url.endswith('swf'):
1393 player_version = self._search_regex(
1394 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1395 'flash player', fatal=False)
1396 player_desc = 'flash player %s' % player_version
1398 player_version = self._search_regex(
1399 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1401 'html5 player', fatal=False)
1402 player_desc = 'html5 player %s' % player_version
1404 parts_sizes = self._signature_cache_id(encrypted_sig)
1405 self.to_screen('{%s} signature length %s, %s' %
1406 (format_id, parts_sizes, player_desc))
1408 signature = self._decrypt_signature(
1409 encrypted_sig, video_id, player_url, age_gate)
1410 url += '&signature=' + signature
1411 if 'ratebypass' not in url:
1412 url += '&ratebypass=yes'
1414 # Some itags are not included in DASH manifest thus corresponding formats will
1415 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1416 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1417 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1418 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1420 'format_id': format_id,
1422 'player_url': player_url,
1423 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1424 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1427 'fps': int_or_none(url_data.get('fps', [None])[0]),
1428 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1430 type_ = url_data.get('type', [None])[0]
1432 type_split = type_.split(';')
1433 kind_ext = type_split[0].split('/')
1434 if len(kind_ext) == 2:
1435 kind, ext = kind_ext
1437 if kind in ('audio', 'video'):
1439 for mobj in re.finditer(
1440 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1441 if mobj.group('key') == 'codecs':
1442 codecs = mobj.group('val')
1445 codecs = codecs.split(',')
1446 if len(codecs) == 2:
1447 acodec, vcodec = codecs[0], codecs[1]
1449 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1454 if format_id in self._formats:
1455 dct.update(self._formats[format_id])
1457 elif video_info.get('hlsvp'):
1458 manifest_url = video_info['hlsvp'][0]
1459 url_map = self._extract_from_m3u8(manifest_url, video_id)
1460 formats = _map_to_format_list(url_map)
1462 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1464 # Look for the DASH manifest
1465 if self._downloader.params.get('youtube_include_dash_manifest', True):
1466 dash_mpd_fatal = True
1467 for dash_manifest_url in dash_mpds:
1470 for df in self._parse_dash_manifest(
1471 video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
1472 # Do not overwrite DASH format found in some previous DASH manifest
1473 if df['format_id'] not in dash_formats:
1474 dash_formats[df['format_id']] = df
1475 # Additional DASH manifests may end up in HTTP Error 403 therefore
1476 # allow them to fail without bug report message if we already have
1477 # some DASH manifest succeeded. This is temporary workaround to reduce
1478 # burst of bug reports until we figure out the reason and whether it
1479 # can be fixed at all.
1480 dash_mpd_fatal = False
1481 except (ExtractorError, KeyError) as e:
1482 self.report_warning(
1483 'Skipping DASH manifest: %r' % e, video_id)
1485 # Remove the formats we found through non-DASH, they
1486 # contain less info and it can be wrong, because we use
1487 # fixed values (for example the resolution). See
1488 # https://github.com/rg3/youtube-dl/issues/5774 for an
1490 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1491 formats.extend(dash_formats.values())
1493 # Check for malformed aspect ratio
1494 stretched_m = re.search(
1495 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1498 ratio = float(stretched_m.group('w')) / float(stretched_m.group('h'))
1500 if f.get('vcodec') != 'none':
1501 f['stretched_ratio'] = ratio
1503 self._sort_formats(formats)
1507 'uploader': video_uploader,
1508 'uploader_id': video_uploader_id,
1509 'upload_date': upload_date,
1510 'title': video_title,
1511 'thumbnail': video_thumbnail,
1512 'description': video_description,
1513 'categories': video_categories,
1515 'subtitles': video_subtitles,
1516 'automatic_captions': automatic_captions,
1517 'duration': video_duration,
1518 'age_limit': 18 if age_gate else 0,
1519 'annotations': video_annotations,
1520 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1521 'view_count': view_count,
1522 'like_count': like_count,
1523 'dislike_count': dislike_count,
1524 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1527 'start_time': start_time,
1528 'end_time': end_time,
1532 class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor):
1533 IE_DESC = 'YouTube.com playlists'
1534 _VALID_URL = r"""(?x)(?:
1539 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1540 \? (?:.*?&)*? (?:p|a|list)=
1544 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1545 # Top tracks, they can also include dots
1550 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1552 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1553 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1554 IE_NAME = 'youtube:playlist'
1556 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1558 'title': 'ytdl test PL',
1559 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1561 'playlist_count': 3,
1563 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1565 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1566 'title': 'YDL_Empty_List',
1568 'playlist_count': 0,
1570 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1571 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1573 'title': '29C3: Not my department',
1574 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1576 'playlist_count': 95,
1578 'note': 'issue #673',
1579 'url': 'PLBB231211A4F62143',
1581 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1582 'id': 'PLBB231211A4F62143',
1584 'playlist_mincount': 26,
1586 'note': 'Large playlist',
1587 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1589 'title': 'Uploads from Cauchemar',
1590 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1592 'playlist_mincount': 799,
1594 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1596 'title': 'YDL_safe_search',
1597 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1599 'playlist_count': 2,
1602 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1603 'playlist_count': 4,
1606 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1609 'note': 'Embedded SWF player',
1610 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1611 'playlist_count': 4,
1614 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1617 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1618 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1620 'title': 'Uploads from Interstellar Movie',
1621 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1623 'playlist_mincout': 21,
1626 def _real_initialize(self):
1629 def _extract_mix(self, playlist_id):
1630 # The mixes are generated from a single video
1631 # the id of the playlist is just 'RD' + video_id
1632 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1633 webpage = self._download_webpage(
1634 url, playlist_id, 'Downloading Youtube mix')
1635 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1637 search_title('playlist-title') or
1638 search_title('title long-title') or
1639 search_title('title'))
1640 title = clean_html(title_span)
1641 ids = orderedSet(re.findall(
1642 r'''(?xs)data-video-username=".*?".*?
1643 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1645 url_results = self._ids_to_results(ids)
1647 return self.playlist_result(url_results, playlist_id, title)
1649 def _extract_playlist(self, playlist_id):
1650 url = self._TEMPLATE_URL % playlist_id
1651 page = self._download_webpage(url, playlist_id)
1653 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1654 match = match.strip()
1655 # Check if the playlist exists or is private
1656 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1657 raise ExtractorError(
1658 'The playlist doesn\'t exist or is private, use --username or '
1659 '--netrc to access it.',
1661 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1662 raise ExtractorError(
1663 'Invalid parameters. Maybe URL is incorrect.',
1665 elif re.match(r'[^<]*Choose your language[^<]*', match):
1668 self.report_warning('Youtube gives an alert message: ' + match)
1670 playlist_title = self._html_search_regex(
1671 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1674 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1676 def _real_extract(self, url):
1677 # Extract playlist id
1678 mobj = re.match(self._VALID_URL, url)
1680 raise ExtractorError('Invalid URL: %s' % url)
1681 playlist_id = mobj.group(1) or mobj.group(2)
1683 # Check if it's a video-specific URL
1684 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1685 if 'v' in query_dict:
1686 video_id = query_dict['v'][0]
1687 if self._downloader.params.get('noplaylist'):
1688 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1689 return self.url_result(video_id, 'Youtube', video_id=video_id)
1691 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1693 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1694 # Mixes require a custom extraction process
1695 return self._extract_mix(playlist_id)
1697 return self._extract_playlist(playlist_id)
1700 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1701 IE_DESC = 'YouTube.com channels'
1702 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1703 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1704 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1705 IE_NAME = 'youtube:channel'
1707 'note': 'paginated channel',
1708 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1709 'playlist_mincount': 91,
1711 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1712 'title': 'Uploads from lex will',
1715 'note': 'Age restricted channel',
1716 # from https://www.youtube.com/user/DeusExOfficial
1717 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1718 'playlist_mincount': 64,
1720 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1721 'title': 'Uploads from Deus Ex',
1725 def _real_extract(self, url):
1726 channel_id = self._match_id(url)
1728 url = self._TEMPLATE_URL % channel_id
1730 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1731 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1732 # otherwise fallback on channel by page extraction
1733 channel_page = self._download_webpage(
1734 url + '?view=57', channel_id,
1735 'Downloading channel page', fatal=False)
1736 if channel_page is False:
1737 channel_playlist_id = False
1739 channel_playlist_id = self._html_search_meta(
1740 'channelId', channel_page, 'channel id', default=None)
1741 if not channel_playlist_id:
1742 channel_playlist_id = self._search_regex(
1743 r'data-(?:channel-external-|yt)id="([^"]+)"',
1744 channel_page, 'channel id', default=None)
1745 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1746 playlist_id = 'UU' + channel_playlist_id[2:]
1747 return self.url_result(
1748 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1750 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1751 autogenerated = re.search(r'''(?x)
1753 channel-header-autogenerated-label|
1754 yt-channel-title-autogenerated
1755 )[^"]*"''', channel_page) is not None
1758 # The videos are contained in a single page
1759 # the ajax pages can't be used, they are empty
1762 video_id, 'Youtube', video_id=video_id,
1763 video_title=video_title)
1764 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1765 return self.playlist_result(entries, channel_id)
1767 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1770 class YoutubeUserIE(YoutubeChannelIE):
1771 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1772 _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_-]+)'
1773 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1774 IE_NAME = 'youtube:user'
1777 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1778 'playlist_mincount': 320,
1780 'title': 'TheLinuxFoundation',
1783 'url': 'ytuser:phihag',
1784 'only_matching': True,
1788 def suitable(cls, url):
1789 # Don't return True if the url can be extracted with other youtube
1790 # extractor, the regex would is too permissive and it would match.
1791 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1792 if any(ie.suitable(url) for ie in other_ies):
1795 return super(YoutubeUserIE, cls).suitable(url)
1798 class YoutubeUserPlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1799 IE_DESC = 'YouTube.com user playlists'
1800 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/user/(?P<id>[^/]+)/playlists'
1801 IE_NAME = 'youtube:user:playlists'
1804 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1805 'playlist_mincount': 4,
1807 'id': 'ThirstForScience',
1808 'title': 'Thirst for Science',
1811 # with "Load more" button
1812 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1813 'playlist_mincount': 70,
1816 'title': 'Игорь Клейнер',
1821 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1822 IE_DESC = 'YouTube.com searches'
1823 # there doesn't appear to be a real limit, for example if you search for
1824 # 'python' you get more than 8.000.000 results
1825 _MAX_RESULTS = float('inf')
1826 IE_NAME = 'youtube:search'
1827 _SEARCH_KEY = 'ytsearch'
1828 _EXTRA_QUERY_ARGS = {}
1831 def _get_n_results(self, query, n):
1832 """Get a specified number of results for a query"""
1837 for pagenum in itertools.count(1):
1839 'search_query': query.encode('utf-8'),
1843 url_query.update(self._EXTRA_QUERY_ARGS)
1844 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1845 data = self._download_json(
1846 result_url, video_id='query "%s"' % query,
1847 note='Downloading page %s' % pagenum,
1848 errnote='Unable to download API page')
1849 html_content = data[1]['body']['content']
1851 if 'class="search-message' in html_content:
1852 raise ExtractorError(
1853 '[youtube] No video results', expected=True)
1855 new_videos = self._ids_to_results(orderedSet(re.findall(
1856 r'href="/watch\?v=(.{11})', html_content)))
1857 videos += new_videos
1858 if not new_videos or len(videos) > limit:
1863 return self.playlist_result(videos, query)
1866 class YoutubeSearchDateIE(YoutubeSearchIE):
1867 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1868 _SEARCH_KEY = 'ytsearchdate'
1869 IE_DESC = 'YouTube.com searches, newest videos first'
1870 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1873 class YoutubeSearchURLIE(InfoExtractor):
1874 IE_DESC = 'YouTube.com search URLs'
1875 IE_NAME = 'youtube:search_url'
1876 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1878 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1879 'playlist_mincount': 5,
1881 'title': 'youtube-dl test video',
1885 def _real_extract(self, url):
1886 mobj = re.match(self._VALID_URL, url)
1887 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
1889 webpage = self._download_webpage(url, query)
1890 result_code = self._search_regex(
1891 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1893 part_codes = re.findall(
1894 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
1896 for part_code in part_codes:
1897 part_title = self._html_search_regex(
1898 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1899 part_url_snippet = self._html_search_regex(
1900 r'(?s)href="([^"]+)"', part_code, 'item URL')
1901 part_url = compat_urlparse.urljoin(
1902 'https://www.youtube.com/', part_url_snippet)
1906 'title': part_title,
1910 '_type': 'playlist',
1916 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
1917 IE_DESC = 'YouTube.com (multi-season) shows'
1918 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1919 IE_NAME = 'youtube:show'
1921 'url': 'https://www.youtube.com/show/airdisasters',
1922 'playlist_mincount': 5,
1924 'id': 'airdisasters',
1925 'title': 'Air Disasters',
1929 def _real_extract(self, url):
1930 playlist_id = self._match_id(url)
1931 return super(YoutubeShowIE, self)._real_extract(
1932 'https://www.youtube.com/show/%s/playlists' % playlist_id)
1935 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1937 Base class for feed extractors
1938 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1940 _LOGIN_REQUIRED = True
1944 return 'youtube:%s' % self._FEED_NAME
1946 def _real_initialize(self):
1949 def _real_extract(self, url):
1950 page = self._download_webpage(
1951 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1953 # The extraction process is the same as for playlists, but the regex
1954 # for the video ids doesn't contain an index
1956 more_widget_html = content_html = page
1957 for page_num in itertools.count(1):
1958 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1960 # 'recommended' feed has infinite 'load more' and each new portion spins
1961 # the same videos in (sometimes) slightly different order, so we'll check
1962 # for unicity and break when portion has no new videos
1963 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1969 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1973 more = self._download_json(
1974 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
1975 'Downloading page #%s' % page_num,
1976 transform_source=uppercase_escape)
1977 content_html = more['content_html']
1978 more_widget_html = more['load_more_widget_html']
1980 return self.playlist_result(
1981 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
1984 class YoutubeWatchLaterIE(YoutubePlaylistIE):
1985 IE_NAME = 'youtube:watchlater'
1986 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1987 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1989 _TESTS = [] # override PlaylistIE tests
1991 def _real_extract(self, url):
1992 return self._extract_playlist('WL')
1995 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
1996 IE_NAME = 'youtube:favorites'
1997 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1998 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1999 _LOGIN_REQUIRED = True
2001 def _real_extract(self, url):
2002 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2003 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2004 return self.url_result(playlist_id, 'YoutubePlaylist')
2007 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2008 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2009 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2010 _FEED_NAME = 'recommended'
2011 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2014 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2015 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2016 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2017 _FEED_NAME = 'subscriptions'
2018 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2021 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2022 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2023 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2024 _FEED_NAME = 'history'
2025 _PLAYLIST_TITLE = 'Youtube History'
2028 class YoutubeTruncatedURLIE(InfoExtractor):
2029 IE_NAME = 'youtube:truncated_url'
2030 IE_DESC = False # Do not list
2031 _VALID_URL = r'''(?x)
2033 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2036 annotation_id=annotation_[^&]+|
2042 attribution_link\?a=[^&]+
2048 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2049 'only_matching': True,
2051 'url': 'http://www.youtube.com/watch?',
2052 'only_matching': True,
2054 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2055 'only_matching': True,
2057 'url': 'https://www.youtube.com/watch?feature=foo',
2058 'only_matching': True,
2060 'url': 'https://www.youtube.com/watch?hl=en-GB',
2061 'only_matching': True,
2063 'url': 'https://www.youtube.com/watch?t=2372',
2064 'only_matching': True,
2067 def _real_extract(self, url):
2068 raise ExtractorError(
2069 'Did you forget to quote the URL? Remember that & is a meta '
2070 'character in most shells, so you want to put the URL in quotes, '
2072 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2073 ' or simply youtube-dl BaW_jenozKc .',
2077 class YoutubeTruncatedIDIE(InfoExtractor):
2078 IE_NAME = 'youtube:truncated_id'
2079 IE_DESC = False # Do not list
2080 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2083 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2084 'only_matching': True,
2087 def _real_extract(self, url):
2088 video_id = self._match_id(url)
2089 raise ExtractorError(
2090 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),