3 from __future__ import unicode_literals
13 from .common import InfoExtractor, SearchInfoExtractor
14 from ..jsinterp import JSInterpreter
15 from ..swfinterp import SWFInterpreter
16 from ..compat import (
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlparse,
32 get_element_by_attribute,
51 class YoutubeBaseInfoExtractor(InfoExtractor):
52 """Provide base functions for Youtube extractors"""
53 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
54 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
55 _NETRC_MACHINE = 'youtube'
56 # If True it will raise an error if no login info is provided
57 _LOGIN_REQUIRED = False
59 def _set_language(self):
61 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
62 # YouTube sets the expire time to about two months
63 expire_time=time.time() + 2 * 30 * 24 * 3600)
65 def _ids_to_results(self, ids):
67 self.url_result(vid_id, 'Youtube', video_id=vid_id)
72 Attempt to log in to YouTube.
73 True is returned if successful or skipped.
74 False is returned if login failed.
76 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
78 (username, password) = self._get_login_info()
79 # No authentication to be performed
81 if self._LOGIN_REQUIRED:
82 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
85 login_page = self._download_webpage(
86 self._LOGIN_URL, None,
87 note='Downloading login page',
88 errnote='unable to fetch login page', fatal=False)
89 if login_page is False:
92 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
93 login_page, 'Login GALX parameter')
97 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
102 'PersistentCookie': 'yes',
104 'bgresponse': 'js_disabled',
105 'checkConnection': '',
106 'checkedDomains': 'youtube',
113 'service': 'youtube',
118 login_data = compat_urllib_parse.urlencode(encode_dict(login_form_strs)).encode('ascii')
120 req = sanitized_Request(self._LOGIN_URL, login_data)
121 login_results = self._download_webpage(
123 note='Logging in', errnote='unable to log in', fatal=False)
124 if login_results is False:
127 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
128 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
131 # TODO add SMS and phone call support - these require making a request and then prompting the user
133 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
134 tfa_code = self._get_tfa_info('2-step verification code')
137 self._downloader.report_warning(
138 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
139 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
142 tfa_code = remove_start(tfa_code, 'G-')
144 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
146 tfa_form_strs.update({
151 tfa_data = compat_urllib_parse.urlencode(encode_dict(tfa_form_strs)).encode('ascii')
153 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
154 tfa_results = self._download_webpage(
156 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
158 if tfa_results is False:
161 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
162 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
164 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
165 self._downloader.report_warning('unable to log in - did the page structure change?')
167 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
168 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
171 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
172 self._downloader.report_warning('unable to log in: bad username or password')
176 def _real_initialize(self):
177 if self._downloader is None:
180 if not self._login():
184 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
185 # Extract entries from page with "Load more" button
186 def _entries(self, page, playlist_id):
187 more_widget_html = content_html = page
188 for page_num in itertools.count(1):
189 for entry in self._process_page(content_html):
192 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
196 more = self._download_json(
197 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
198 'Downloading page #%s' % page_num,
199 transform_source=uppercase_escape)
200 content_html = more['content_html']
201 if not content_html.strip():
202 # Some webpages show a "Load more" button but they don't
205 more_widget_html = more['load_more_widget_html']
208 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
209 def _process_page(self, content):
210 for video_id, video_title in self.extract_videos_from_page(content):
211 yield self.url_result(video_id, 'Youtube', video_id, video_title)
213 def extract_videos_from_page(self, page):
216 for mobj in re.finditer(self._VIDEO_RE, page):
217 # The link with index 0 is not the first video of the playlist (not sure if still actual)
218 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
220 video_id = mobj.group('id')
221 video_title = unescapeHTML(mobj.group('title'))
223 video_title = video_title.strip()
225 idx = ids_in_page.index(video_id)
226 if video_title and not titles_in_page[idx]:
227 titles_in_page[idx] = video_title
229 ids_in_page.append(video_id)
230 titles_in_page.append(video_title)
231 return zip(ids_in_page, titles_in_page)
234 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
235 def _process_page(self, content):
236 for playlist_id in orderedSet(re.findall(r'href="/?playlist\?list=([0-9A-Za-z-_]{10,})"', content)):
237 yield self.url_result(
238 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
240 def _real_extract(self, url):
241 playlist_id = self._match_id(url)
242 webpage = self._download_webpage(url, playlist_id)
243 title = self._og_search_title(webpage, fatal=False)
244 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
247 class YoutubeIE(YoutubeBaseInfoExtractor):
248 IE_DESC = 'YouTube.com'
249 _VALID_URL = r"""(?x)^
251 (?:https?://|//) # http(s):// or protocol-independent URL
252 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
253 (?:www\.)?deturl\.com/www\.youtube\.com/|
254 (?:www\.)?pwnyoutube\.com/|
255 (?:www\.)?yourepeat\.com/|
256 tube\.majestyc\.net/|
257 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
258 (?:.*?\#/)? # handle anchor (#/) redirect urls
259 (?: # the various things that can precede the ID:
260 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
261 |(?: # or the v= param in all its forms
262 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
263 (?:\?|\#!?) # the params delimiter ? or # or #!
264 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
269 youtu\.be| # just youtu.be/xxxx
270 vid\.plus # or vid.plus/xxxx
272 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
274 )? # all until now is optional -> you can pass the naked ID
275 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
276 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
277 (?(1).+)? # if we found the ID, everything can follow
279 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
281 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
282 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
283 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
284 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
285 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
286 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
287 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
288 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
289 '36': {'ext': '3gp', 'width': 320, 'height': 240, 'acodec': 'aac', 'abr': 32, 'vcodec': 'mp4v'},
290 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
291 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
292 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
293 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
294 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
295 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
296 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
297 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
301 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
302 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
303 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
304 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
305 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
306 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
307 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
309 # Apple HTTP Live Streaming
310 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
311 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
312 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
313 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
314 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
315 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
316 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
319 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
320 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
321 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
322 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
323 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
324 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
325 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
326 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
327 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
328 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
329 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
332 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
333 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
334 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
337 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
338 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
339 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
340 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
341 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
342 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
343 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
344 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
345 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
346 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
347 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
348 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
349 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
350 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
351 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
352 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
353 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
354 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
355 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
356 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
357 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
358 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
361 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
362 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
364 # Dash webm audio with opus inside
365 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
366 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
367 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
370 '_rtmp': {'protocol': 'rtmp'},
372 _SUBTITLE_FORMATS = ('ttml', 'vtt')
377 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
381 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
382 'uploader': 'Philipp Hagemeister',
383 'uploader_id': 'phihag',
384 'upload_date': '20121002',
385 '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 .',
386 'categories': ['Science & Technology'],
387 'tags': ['youtube-dl'],
389 'dislike_count': int,
395 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
396 'note': 'Test generic use_cipher_signature video (#897)',
400 'upload_date': '20120506',
401 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
402 'alt_title': 'I Love It (feat. Charli XCX)',
403 'description': 'md5:782e8651347686cba06e58f71ab51773',
404 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
405 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
406 'iconic ep', 'iconic', 'love', 'it'],
407 'uploader': 'Icona Pop',
408 'uploader_id': 'IconaPop',
409 'creator': 'Icona Pop',
413 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
414 'note': 'Test VEVO video with age protection (#956)',
418 'upload_date': '20130703',
419 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
420 'alt_title': 'Tunnel Vision',
421 'description': 'md5:64249768eec3bc4276236606ea996373',
422 'uploader': 'justintimberlakeVEVO',
423 'uploader_id': 'justintimberlakeVEVO',
424 'creator': 'Justin Timberlake',
429 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
430 'note': 'Embed-only video (#1746)',
434 'upload_date': '20120608',
435 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
436 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
437 'uploader': 'SET India',
438 'uploader_id': 'setindia',
443 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY',
444 'note': 'Use the first video ID in the URL',
448 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
449 'uploader': 'Philipp Hagemeister',
450 'uploader_id': 'phihag',
451 'upload_date': '20121002',
452 '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 .',
453 'categories': ['Science & Technology'],
454 'tags': ['youtube-dl'],
456 'dislike_count': int,
459 'skip_download': True,
463 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
464 'note': '256k DASH audio (format 141) via DASH manifest',
468 'upload_date': '20121002',
469 'uploader_id': '8KVIDEO',
471 'uploader': '8KVIDEO',
472 'title': 'UHDTV TEST 8K VIDEO.mp4'
475 'youtube_include_dash_manifest': True,
479 # DASH manifest with encrypted signature
481 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
485 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
486 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
487 'uploader': 'AfrojackVEVO',
488 'uploader_id': 'AfrojackVEVO',
489 'upload_date': '20131011',
492 'youtube_include_dash_manifest': True,
496 # JS player signature function name containing $
498 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
502 'title': 'Taylor Swift - Shake It Off',
503 'alt_title': 'Shake It Off',
504 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
505 'uploader': 'TaylorSwiftVEVO',
506 'uploader_id': 'TaylorSwiftVEVO',
507 'upload_date': '20140818',
508 'creator': 'Taylor Swift',
511 'youtube_include_dash_manifest': True,
517 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
521 'upload_date': '20100909',
522 'uploader': 'The Amazing Atheist',
523 'uploader_id': 'TheAmazingAtheist',
524 'title': 'Burning Everyone\'s Koran',
525 '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',
528 # Normal age-gate video (No vevo, embed allowed)
530 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
534 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
535 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
536 'uploader': 'The Witcher',
537 'uploader_id': 'WitcherGame',
538 'upload_date': '20140605',
542 # Age-gate video with encrypted signature
544 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
548 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
549 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
550 'uploader': 'LloydVEVO',
551 'uploader_id': 'LloydVEVO',
552 'upload_date': '20110629',
556 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
558 'url': '__2ABJjxzNo',
562 'upload_date': '20100430',
563 'uploader_id': 'deadmau5',
564 'creator': 'deadmau5',
565 'description': 'md5:12c56784b8032162bb936a5f76d55360',
566 'uploader': 'deadmau5',
567 'title': 'Deadmau5 - Some Chords (HD)',
568 'alt_title': 'Some Chords',
570 'expected_warnings': [
571 'DASH manifest missing',
574 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
576 'url': 'lqQg6PlCWgI',
580 'upload_date': '20150827',
581 'uploader_id': 'olympic',
582 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
583 'uploader': 'Olympics',
584 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
587 'skip_download': 'requires avconv',
592 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
596 'stretched_ratio': 16 / 9.,
597 'upload_date': '20110310',
598 'uploader_id': 'AllenMeow',
599 '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',
601 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
604 # url_encoded_fmt_stream_map is empty string
606 'url': 'qEJwOuvDf7I',
610 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
612 'upload_date': '20150404',
613 'uploader_id': 'spbelect',
614 'uploader': 'Наблюдатели Петербурга',
617 'skip_download': 'requires avconv',
619 'skip': 'This live event has ended.',
621 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
623 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
627 'title': 'md5:7b81415841e02ecd4313668cde88737a',
628 'description': 'md5:116377fd2963b81ec4ce64b542173306',
629 'upload_date': '20150625',
630 'uploader_id': 'dorappi2000',
631 'uploader': 'dorappi2000',
632 'formats': 'mincount:33',
635 # DASH manifest with segment_list
637 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
638 'md5': '8ce563a1d667b599d21064e982ab9e31',
642 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
643 'uploader': 'Airtek',
644 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
645 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
646 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
649 'youtube_include_dash_manifest': True,
650 'format': '135', # bestvideo
654 # Multifeed videos (multiple cameras), URL is for Main Camera
655 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
658 'title': 'teamPGP: Rocket League Noob Stream',
659 'description': 'md5:dc7872fb300e143831327f1bae3af010',
665 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
666 'description': 'md5:dc7872fb300e143831327f1bae3af010',
667 'upload_date': '20150721',
668 'uploader': 'Beer Games Beer',
669 'uploader_id': 'beergamesbeer',
675 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
676 'description': 'md5:dc7872fb300e143831327f1bae3af010',
677 'upload_date': '20150721',
678 'uploader': 'Beer Games Beer',
679 'uploader_id': 'beergamesbeer',
685 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
686 'description': 'md5:dc7872fb300e143831327f1bae3af010',
687 'upload_date': '20150721',
688 'uploader': 'Beer Games Beer',
689 'uploader_id': 'beergamesbeer',
695 'title': 'teamPGP: Rocket League Noob Stream (zim)',
696 'description': 'md5:dc7872fb300e143831327f1bae3af010',
697 'upload_date': '20150721',
698 'uploader': 'Beer Games Beer',
699 'uploader_id': 'beergamesbeer',
703 'skip_download': True,
707 'url': 'http://vid.plus/FlRa-iH7PGw',
708 'only_matching': True,
711 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
712 # Also tests cut-off URL expansion in video description (see
713 # https://github.com/rg3/youtube-dl/issues/1892,
714 # https://github.com/rg3/youtube-dl/issues/8164)
715 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
719 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
720 'alt_title': 'Dark Walk',
721 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
722 'upload_date': '20151119',
723 'uploader_id': 'IronSoulElf',
724 'uploader': 'IronSoulElf',
725 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
728 'skip_download': True,
732 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
733 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
734 'only_matching': True,
737 # Video with yt:stretch=17:0
738 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
742 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
743 'description': 'md5:ee18a25c350637c8faff806845bddee9',
744 'upload_date': '20151107',
745 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
746 'uploader': 'CH GAMER DROID',
749 'skip_download': True,
753 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
754 'only_matching': True,
758 def __init__(self, *args, **kwargs):
759 super(YoutubeIE, self).__init__(*args, **kwargs)
760 self._player_cache = {}
762 def report_video_info_webpage_download(self, video_id):
763 """Report attempt to download video info webpage."""
764 self.to_screen('%s: Downloading video info webpage' % video_id)
766 def report_information_extraction(self, video_id):
767 """Report attempt to extract video information."""
768 self.to_screen('%s: Extracting video information' % video_id)
770 def report_unavailable_format(self, video_id, format):
771 """Report extracted video URL."""
772 self.to_screen('%s: Format %s not available' % (video_id, format))
774 def report_rtmp_download(self):
775 """Indicate the download will use the RTMP protocol."""
776 self.to_screen('RTMP download detected')
778 def _signature_cache_id(self, example_sig):
779 """ Return a string representation of a signature """
780 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
782 def _extract_signature_function(self, video_id, player_url, example_sig):
784 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
787 raise ExtractorError('Cannot identify player %r' % player_url)
788 player_type = id_m.group('ext')
789 player_id = id_m.group('id')
791 # Read from filesystem cache
792 func_id = '%s_%s_%s' % (
793 player_type, player_id, self._signature_cache_id(example_sig))
794 assert os.path.basename(func_id) == func_id
796 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
797 if cache_spec is not None:
798 return lambda s: ''.join(s[i] for i in cache_spec)
801 'Downloading player %s' % player_url
802 if self._downloader.params.get('verbose') else
803 'Downloading %s player %s' % (player_type, player_id)
805 if player_type == 'js':
806 code = self._download_webpage(
807 player_url, video_id,
809 errnote='Download of %s failed' % player_url)
810 res = self._parse_sig_js(code)
811 elif player_type == 'swf':
812 urlh = self._request_webpage(
813 player_url, video_id,
815 errnote='Download of %s failed' % player_url)
817 res = self._parse_sig_swf(code)
819 assert False, 'Invalid player type %r' % player_type
821 test_string = ''.join(map(compat_chr, range(len(example_sig))))
822 cache_res = res(test_string)
823 cache_spec = [ord(c) for c in cache_res]
825 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
828 def _print_sig_code(self, func, example_sig):
829 def gen_sig_code(idxs):
830 def _genslice(start, end, step):
831 starts = '' if start == 0 else str(start)
832 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
833 steps = '' if step == 1 else (':%d' % step)
834 return 's[%s%s%s]' % (starts, ends, steps)
837 # Quelch pyflakes warnings - start will be set when step is set
838 start = '(Never used)'
839 for i, prev in zip(idxs[1:], idxs[:-1]):
843 yield _genslice(start, prev, step)
846 if i - prev in [-1, 1]:
855 yield _genslice(start, i, step)
857 test_string = ''.join(map(compat_chr, range(len(example_sig))))
858 cache_res = func(test_string)
859 cache_spec = [ord(c) for c in cache_res]
860 expr_code = ' + '.join(gen_sig_code(cache_spec))
861 signature_id_tuple = '(%s)' % (
862 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
863 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
864 ' return %s\n') % (signature_id_tuple, expr_code)
865 self.to_screen('Extracted signature function:\n' + code)
867 def _parse_sig_js(self, jscode):
868 funcname = self._search_regex(
869 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
870 'Initial JS player signature function name')
872 jsi = JSInterpreter(jscode)
873 initial_function = jsi.extract_function(funcname)
874 return lambda s: initial_function([s])
876 def _parse_sig_swf(self, file_contents):
877 swfi = SWFInterpreter(file_contents)
878 TARGET_CLASSNAME = 'SignatureDecipher'
879 searched_class = swfi.extract_class(TARGET_CLASSNAME)
880 initial_function = swfi.extract_function(searched_class, 'decipher')
881 return lambda s: initial_function([s])
883 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
884 """Turn the encrypted s field into a working signature"""
886 if player_url is None:
887 raise ExtractorError('Cannot decrypt signature without player_url')
889 if player_url.startswith('//'):
890 player_url = 'https:' + player_url
892 player_id = (player_url, self._signature_cache_id(s))
893 if player_id not in self._player_cache:
894 func = self._extract_signature_function(
895 video_id, player_url, s
897 self._player_cache[player_id] = func
898 func = self._player_cache[player_id]
899 if self._downloader.params.get('youtube_print_sig_code'):
900 self._print_sig_code(func, s)
902 except Exception as e:
903 tb = traceback.format_exc()
904 raise ExtractorError(
905 'Signature extraction failed: ' + tb, cause=e)
907 def _get_subtitles(self, video_id, webpage):
909 subs_doc = self._download_xml(
910 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
911 video_id, note=False)
912 except ExtractorError as err:
913 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
917 for track in subs_doc.findall('track'):
918 lang = track.attrib['lang_code']
919 if lang in sub_lang_list:
922 for ext in self._SUBTITLE_FORMATS:
923 params = compat_urllib_parse.urlencode({
927 'name': track.attrib['name'].encode('utf-8'),
930 'url': 'https://www.youtube.com/api/timedtext?' + params,
933 sub_lang_list[lang] = sub_formats
934 if not sub_lang_list:
935 self._downloader.report_warning('video doesn\'t have subtitles')
939 def _get_ytplayer_config(self, video_id, webpage):
941 # User data may contain arbitrary character sequences that may affect
942 # JSON extraction with regex, e.g. when '};' is contained the second
943 # regex won't capture the whole JSON. Yet working around by trying more
944 # concrete regex first keeping in mind proper quoted string handling
945 # to be implemented in future that will replace this workaround (see
946 # https://github.com/rg3/youtube-dl/issues/7468,
947 # https://github.com/rg3/youtube-dl/pull/7599)
948 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
949 r';ytplayer\.config\s*=\s*({.+?});',
951 config = self._search_regex(
952 patterns, webpage, 'ytplayer.config', default=None)
954 return self._parse_json(
955 uppercase_escape(config), video_id, fatal=False)
957 def _get_automatic_captions(self, video_id, webpage):
958 """We need the webpage for getting the captions url, pass it as an
959 argument to speed up the process."""
960 self.to_screen('%s: Looking for automatic captions' % video_id)
961 player_config = self._get_ytplayer_config(video_id, webpage)
962 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
963 if not player_config:
964 self._downloader.report_warning(err_msg)
967 args = player_config['args']
968 caption_url = args['ttsurl']
970 self._downloader.report_warning(err_msg)
972 timestamp = args['timestamp']
973 # We get the available subtitles
974 list_params = compat_urllib_parse.urlencode({
979 list_url = caption_url + '&' + list_params
980 caption_list = self._download_xml(list_url, video_id)
981 original_lang_node = caption_list.find('track')
982 if original_lang_node is None:
983 self._downloader.report_warning('Video doesn\'t have automatic captions')
985 original_lang = original_lang_node.attrib['lang_code']
986 caption_kind = original_lang_node.attrib.get('kind', '')
989 for lang_node in caption_list.findall('target'):
990 sub_lang = lang_node.attrib['lang_code']
992 for ext in self._SUBTITLE_FORMATS:
993 params = compat_urllib_parse.urlencode({
994 'lang': original_lang,
998 'kind': caption_kind,
1000 sub_formats.append({
1001 'url': caption_url + '&' + params,
1004 sub_lang_list[sub_lang] = sub_formats
1005 return sub_lang_list
1006 # An extractor error can be raise by the download process if there are
1007 # no automatic captions but there are subtitles
1008 except (KeyError, ExtractorError):
1009 self._downloader.report_warning(err_msg)
1013 def extract_id(cls, url):
1014 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1016 raise ExtractorError('Invalid URL: %s' % url)
1017 video_id = mobj.group(2)
1020 def _extract_from_m3u8(self, manifest_url, video_id):
1023 def _get_urls(_manifest):
1024 lines = _manifest.split('\n')
1025 urls = filter(lambda l: l and not l.startswith('#'),
1028 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1029 formats_urls = _get_urls(manifest)
1030 for format_url in formats_urls:
1031 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1032 url_map[itag] = format_url
1035 def _extract_annotations(self, video_id):
1036 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1037 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1039 def _real_extract(self, url):
1040 url, smuggled_data = unsmuggle_url(url, {})
1043 'http' if self._downloader.params.get('prefer_insecure', False)
1048 parsed_url = compat_urllib_parse_urlparse(url)
1049 for component in [parsed_url.fragment, parsed_url.query]:
1050 query = compat_parse_qs(component)
1051 if start_time is None and 't' in query:
1052 start_time = parse_duration(query['t'][0])
1053 if start_time is None and 'start' in query:
1054 start_time = parse_duration(query['start'][0])
1055 if end_time is None and 'end' in query:
1056 end_time = parse_duration(query['end'][0])
1058 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1059 mobj = re.search(self._NEXT_URL_RE, url)
1061 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1062 video_id = self.extract_id(url)
1065 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1066 video_webpage = self._download_webpage(url, video_id)
1068 # Attempt to extract SWF player URL
1069 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1070 if mobj is not None:
1071 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1077 def add_dash_mpd(video_info):
1078 dash_mpd = video_info.get('dashmpd')
1079 if dash_mpd and dash_mpd[0] not in dash_mpds:
1080 dash_mpds.append(dash_mpd[0])
1083 embed_webpage = None
1085 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1087 # We simulate the access to the video from www.youtube.com/v/{video_id}
1088 # this can be viewed without login into Youtube
1089 url = proto + '://www.youtube.com/embed/%s' % video_id
1090 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1091 data = compat_urllib_parse.urlencode({
1092 'video_id': video_id,
1093 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1094 'sts': self._search_regex(
1095 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1097 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1098 video_info_webpage = self._download_webpage(
1099 video_info_url, video_id,
1100 note='Refetching age-gated info webpage',
1101 errnote='unable to download video info webpage')
1102 video_info = compat_parse_qs(video_info_webpage)
1103 add_dash_mpd(video_info)
1107 # Try looking directly into the video webpage
1108 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1110 args = ytplayer_config['args']
1111 if args.get('url_encoded_fmt_stream_map'):
1112 # Convert to the same format returned by compat_parse_qs
1113 video_info = dict((k, [v]) for k, v in args.items())
1114 add_dash_mpd(video_info)
1115 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1117 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1118 # We also try looking in get_video_info since it may contain different dashmpd
1119 # URL that points to a DASH manifest with possibly different itag set (some itags
1120 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1121 # manifest pointed by get_video_info's dashmpd).
1122 # The general idea is to take a union of itags of both DASH manifests (for example
1123 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1124 self.report_video_info_webpage_download(video_id)
1125 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1127 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1128 % (proto, video_id, el_type))
1129 video_info_webpage = self._download_webpage(
1131 video_id, note=False,
1132 errnote='unable to download video info webpage')
1133 get_video_info = compat_parse_qs(video_info_webpage)
1134 if get_video_info.get('use_cipher_signature') != ['True']:
1135 add_dash_mpd(get_video_info)
1137 video_info = get_video_info
1138 if 'token' in get_video_info:
1139 # Different get_video_info requests may report different results, e.g.
1140 # some may report video unavailability, but some may serve it without
1141 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1142 # the original webpage as well as el=info and el=embedded get_video_info
1143 # requests report video unavailability due to geo restriction while
1144 # el=detailpage succeeds and returns valid data). This is probably
1145 # due to YouTube measures against IP ranges of hosting providers.
1146 # Working around by preferring the first succeeded video_info containing
1147 # the token if no such video_info yet was found.
1148 if 'token' not in video_info:
1149 video_info = get_video_info
1151 if 'token' not in video_info:
1152 if 'reason' in video_info:
1153 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1154 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1156 raise ExtractorError('YouTube said: This video is available in %s only' % (
1157 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1159 raise ExtractorError(
1160 'YouTube said: %s' % video_info['reason'][0],
1161 expected=True, video_id=video_id)
1163 raise ExtractorError(
1164 '"token" parameter not in video info for unknown reason',
1168 if 'title' in video_info:
1169 video_title = video_info['title'][0]
1171 self._downloader.report_warning('Unable to extract video title')
1175 video_description = get_element_by_id("eow-description", video_webpage)
1176 if video_description:
1177 video_description = re.sub(r'''(?x)
1179 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1180 (?:title|href)="([^"]+)"\s+
1181 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1182 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1185 ''', r'\1', video_description)
1186 video_description = clean_html(video_description)
1188 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1190 video_description = unescapeHTML(fd_mobj.group(1))
1192 video_description = ''
1194 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1195 if not self._downloader.params.get('noplaylist'):
1198 multifeed_metadata_list = compat_urllib_parse_unquote_plus(video_info['multifeed_metadata_list'][0])
1199 for feed in multifeed_metadata_list.split(','):
1200 feed_data = compat_parse_qs(feed)
1202 '_type': 'url_transparent',
1203 'ie_key': 'Youtube',
1205 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1206 {'force_singlefeed': True}),
1207 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1209 feed_ids.append(feed_data['id'][0])
1211 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1212 % (', '.join(feed_ids), video_id))
1213 return self.playlist_result(entries, video_id, video_title, video_description)
1214 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1216 if 'view_count' in video_info:
1217 view_count = int(video_info['view_count'][0])
1221 # Check for "rental" videos
1222 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1223 raise ExtractorError('"rental" videos not supported')
1225 # Start extracting information
1226 self.report_information_extraction(video_id)
1229 if 'author' not in video_info:
1230 raise ExtractorError('Unable to extract uploader name')
1231 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1234 video_uploader_id = None
1235 mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
1236 if mobj is not None:
1237 video_uploader_id = mobj.group(1)
1239 self._downloader.report_warning('unable to extract uploader nickname')
1242 # We try first to get a high quality image:
1243 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1244 video_webpage, re.DOTALL)
1245 if m_thumb is not None:
1246 video_thumbnail = m_thumb.group(1)
1247 elif 'thumbnail_url' not in video_info:
1248 self._downloader.report_warning('unable to extract video thumbnail')
1249 video_thumbnail = None
1250 else: # don't panic if we can't find it
1251 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1254 upload_date = self._html_search_meta(
1255 'datePublished', video_webpage, 'upload date', default=None)
1257 upload_date = self._search_regex(
1258 [r'(?s)id="eow-date.*?>(.*?)</span>',
1259 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1260 video_webpage, 'upload date', default=None)
1262 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1263 upload_date = unified_strdate(upload_date)
1265 m_music = re.search(
1266 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1269 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1270 video_creator = clean_html(m_music.group('creator'))
1272 video_alt_title = video_creator = None
1274 m_cat_container = self._search_regex(
1275 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1276 video_webpage, 'categories', default=None)
1278 category = self._html_search_regex(
1279 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1281 video_categories = None if category is None else [category]
1283 video_categories = None
1286 unescapeHTML(m.group('content'))
1287 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1289 def _extract_count(count_name):
1290 return str_to_int(self._search_regex(
1291 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1292 % re.escape(count_name),
1293 video_webpage, count_name, default=None))
1295 like_count = _extract_count('like')
1296 dislike_count = _extract_count('dislike')
1299 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1300 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1302 if 'length_seconds' not in video_info:
1303 self._downloader.report_warning('unable to extract video duration')
1304 video_duration = None
1306 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1309 video_annotations = None
1310 if self._downloader.params.get('writeannotations', False):
1311 video_annotations = self._extract_annotations(video_id)
1313 def _map_to_format_list(urlmap):
1315 for itag, video_real_url in urlmap.items():
1318 'url': video_real_url,
1319 'player_url': player_url,
1321 if itag in self._formats:
1322 dct.update(self._formats[itag])
1326 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1327 self.report_rtmp_download()
1329 'format_id': '_rtmp',
1331 'url': video_info['conn'][0],
1332 'player_url': player_url,
1334 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1335 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1336 if 'rtmpe%3Dyes' in encoded_url_map:
1337 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1339 for url_data_str in encoded_url_map.split(','):
1340 url_data = compat_parse_qs(url_data_str)
1341 if 'itag' not in url_data or 'url' not in url_data:
1343 format_id = url_data['itag'][0]
1344 url = url_data['url'][0]
1346 if 'sig' in url_data:
1347 url += '&signature=' + url_data['sig'][0]
1348 elif 's' in url_data:
1349 encrypted_sig = url_data['s'][0]
1350 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1352 jsplayer_url_json = self._search_regex(
1354 embed_webpage if age_gate else video_webpage,
1355 'JS player URL (1)', default=None)
1356 if not jsplayer_url_json and not age_gate:
1357 # We need the embed website after all
1358 if embed_webpage is None:
1359 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1360 embed_webpage = self._download_webpage(
1361 embed_url, video_id, 'Downloading embed webpage')
1362 jsplayer_url_json = self._search_regex(
1363 ASSETS_RE, embed_webpage, 'JS player URL')
1365 player_url = json.loads(jsplayer_url_json)
1366 if player_url is None:
1367 player_url_json = self._search_regex(
1368 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1369 video_webpage, 'age gate player URL')
1370 player_url = json.loads(player_url_json)
1372 if self._downloader.params.get('verbose'):
1373 if player_url is None:
1374 player_version = 'unknown'
1375 player_desc = 'unknown'
1377 if player_url.endswith('swf'):
1378 player_version = self._search_regex(
1379 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1380 'flash player', fatal=False)
1381 player_desc = 'flash player %s' % player_version
1383 player_version = self._search_regex(
1384 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1386 'html5 player', fatal=False)
1387 player_desc = 'html5 player %s' % player_version
1389 parts_sizes = self._signature_cache_id(encrypted_sig)
1390 self.to_screen('{%s} signature length %s, %s' %
1391 (format_id, parts_sizes, player_desc))
1393 signature = self._decrypt_signature(
1394 encrypted_sig, video_id, player_url, age_gate)
1395 url += '&signature=' + signature
1396 if 'ratebypass' not in url:
1397 url += '&ratebypass=yes'
1400 'format_id': format_id,
1402 'player_url': player_url,
1404 if format_id in self._formats:
1405 dct.update(self._formats[format_id])
1407 # Some itags are not included in DASH manifest thus corresponding formats will
1408 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1409 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1410 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1411 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1414 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1415 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1418 'fps': int_or_none(url_data.get('fps', [None])[0]),
1419 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1421 for key, value in more_fields.items():
1424 type_ = url_data.get('type', [None])[0]
1426 type_split = type_.split(';')
1427 kind_ext = type_split[0].split('/')
1428 if len(kind_ext) == 2:
1430 dct['ext'] = mimetype2ext(type_split[0])
1431 if kind in ('audio', 'video'):
1433 for mobj in re.finditer(
1434 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1435 if mobj.group('key') == 'codecs':
1436 codecs = mobj.group('val')
1439 codecs = codecs.split(',')
1440 if len(codecs) == 2:
1441 acodec, vcodec = codecs[1], codecs[0]
1443 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1449 elif video_info.get('hlsvp'):
1450 manifest_url = video_info['hlsvp'][0]
1451 url_map = self._extract_from_m3u8(manifest_url, video_id)
1452 formats = _map_to_format_list(url_map)
1453 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1454 for a_format in formats:
1455 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1457 unavailable_message = self._html_search_regex(
1458 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1459 video_webpage, 'unavailable message', default=None)
1460 if unavailable_message:
1461 raise ExtractorError(unavailable_message, expected=True)
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 mpd_url in dash_mpds:
1470 def decrypt_sig(mobj):
1472 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1473 return '/signature/%s' % dec_s
1475 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1477 for df in self._extract_mpd_formats(
1478 mpd_url, video_id, fatal=dash_mpd_fatal,
1479 formats_dict=self._formats):
1480 # Do not overwrite DASH format found in some previous DASH manifest
1481 if df['format_id'] not in dash_formats:
1482 dash_formats[df['format_id']] = df
1483 # Additional DASH manifests may end up in HTTP Error 403 therefore
1484 # allow them to fail without bug report message if we already have
1485 # some DASH manifest succeeded. This is temporary workaround to reduce
1486 # burst of bug reports until we figure out the reason and whether it
1487 # can be fixed at all.
1488 dash_mpd_fatal = False
1489 except (ExtractorError, KeyError) as e:
1490 self.report_warning(
1491 'Skipping DASH manifest: %r' % e, video_id)
1493 # Remove the formats we found through non-DASH, they
1494 # contain less info and it can be wrong, because we use
1495 # fixed values (for example the resolution). See
1496 # https://github.com/rg3/youtube-dl/issues/5774 for an
1498 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1499 formats.extend(dash_formats.values())
1501 # Check for malformed aspect ratio
1502 stretched_m = re.search(
1503 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1506 w = float(stretched_m.group('w'))
1507 h = float(stretched_m.group('h'))
1508 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1509 # We will only process correct ratios.
1513 if f.get('vcodec') != 'none':
1514 f['stretched_ratio'] = ratio
1516 self._sort_formats(formats)
1520 'uploader': video_uploader,
1521 'uploader_id': video_uploader_id,
1522 'upload_date': upload_date,
1523 'creator': video_creator,
1524 'title': video_title,
1525 'alt_title': video_alt_title,
1526 'thumbnail': video_thumbnail,
1527 'description': video_description,
1528 'categories': video_categories,
1530 'subtitles': video_subtitles,
1531 'automatic_captions': automatic_captions,
1532 'duration': video_duration,
1533 'age_limit': 18 if age_gate else 0,
1534 'annotations': video_annotations,
1535 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1536 'view_count': view_count,
1537 'like_count': like_count,
1538 'dislike_count': dislike_count,
1539 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1542 'start_time': start_time,
1543 'end_time': end_time,
1547 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1548 IE_DESC = 'YouTube.com playlists'
1549 _VALID_URL = r"""(?x)(?:
1554 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1555 \? (?:.*?[&;])*? (?:p|a|list)=
1559 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1560 # Top tracks, they can also include dots
1565 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1567 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1568 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1569 IE_NAME = 'youtube:playlist'
1571 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1573 'title': 'ytdl test PL',
1574 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1576 'playlist_count': 3,
1578 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1580 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1581 'title': 'YDL_Empty_List',
1583 'playlist_count': 0,
1585 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1586 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1588 'title': '29C3: Not my department',
1589 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1591 'playlist_count': 95,
1593 'note': 'issue #673',
1594 'url': 'PLBB231211A4F62143',
1596 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1597 'id': 'PLBB231211A4F62143',
1599 'playlist_mincount': 26,
1601 'note': 'Large playlist',
1602 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1604 'title': 'Uploads from Cauchemar',
1605 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1607 'playlist_mincount': 799,
1609 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1611 'title': 'YDL_safe_search',
1612 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1614 'playlist_count': 2,
1617 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1618 'playlist_count': 4,
1621 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1624 'note': 'Embedded SWF player',
1625 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1626 'playlist_count': 4,
1629 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1632 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1633 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1635 'title': 'Uploads from Interstellar Movie',
1636 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1638 'playlist_mincout': 21,
1641 def _real_initialize(self):
1644 def _extract_mix(self, playlist_id):
1645 # The mixes are generated from a single video
1646 # the id of the playlist is just 'RD' + video_id
1647 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1648 webpage = self._download_webpage(
1649 url, playlist_id, 'Downloading Youtube mix')
1650 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1652 search_title('playlist-title') or
1653 search_title('title long-title') or
1654 search_title('title'))
1655 title = clean_html(title_span)
1656 ids = orderedSet(re.findall(
1657 r'''(?xs)data-video-username=".*?".*?
1658 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1660 url_results = self._ids_to_results(ids)
1662 return self.playlist_result(url_results, playlist_id, title)
1664 def _extract_playlist(self, playlist_id):
1665 url = self._TEMPLATE_URL % playlist_id
1666 page = self._download_webpage(url, playlist_id)
1668 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1669 match = match.strip()
1670 # Check if the playlist exists or is private
1671 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1672 raise ExtractorError(
1673 'The playlist doesn\'t exist or is private, use --username or '
1674 '--netrc to access it.',
1676 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1677 raise ExtractorError(
1678 'Invalid parameters. Maybe URL is incorrect.',
1680 elif re.match(r'[^<]*Choose your language[^<]*', match):
1683 self.report_warning('Youtube gives an alert message: ' + match)
1685 playlist_title = self._html_search_regex(
1686 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1689 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1691 def _real_extract(self, url):
1692 # Extract playlist id
1693 mobj = re.match(self._VALID_URL, url)
1695 raise ExtractorError('Invalid URL: %s' % url)
1696 playlist_id = mobj.group(1) or mobj.group(2)
1698 # Check if it's a video-specific URL
1699 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1700 if 'v' in query_dict:
1701 video_id = query_dict['v'][0]
1702 if self._downloader.params.get('noplaylist'):
1703 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1704 return self.url_result(video_id, 'Youtube', video_id=video_id)
1706 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1708 if playlist_id.startswith('RD') or playlist_id.startswith('UL'):
1709 # Mixes require a custom extraction process
1710 return self._extract_mix(playlist_id)
1712 return self._extract_playlist(playlist_id)
1715 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1716 IE_DESC = 'YouTube.com channels'
1717 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1718 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1719 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1720 IE_NAME = 'youtube:channel'
1722 'note': 'paginated channel',
1723 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1724 'playlist_mincount': 91,
1726 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1727 'title': 'Uploads from lex will',
1730 'note': 'Age restricted channel',
1731 # from https://www.youtube.com/user/DeusExOfficial
1732 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1733 'playlist_mincount': 64,
1735 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1736 'title': 'Uploads from Deus Ex',
1741 def suitable(cls, url):
1742 return False if YoutubePlaylistsIE.suitable(url) else super(YoutubeChannelIE, cls).suitable(url)
1744 def _real_extract(self, url):
1745 channel_id = self._match_id(url)
1747 url = self._TEMPLATE_URL % channel_id
1749 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1750 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1751 # otherwise fallback on channel by page extraction
1752 channel_page = self._download_webpage(
1753 url + '?view=57', channel_id,
1754 'Downloading channel page', fatal=False)
1755 if channel_page is False:
1756 channel_playlist_id = False
1758 channel_playlist_id = self._html_search_meta(
1759 'channelId', channel_page, 'channel id', default=None)
1760 if not channel_playlist_id:
1761 channel_playlist_id = self._search_regex(
1762 r'data-(?:channel-external-|yt)id="([^"]+)"',
1763 channel_page, 'channel id', default=None)
1764 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1765 playlist_id = 'UU' + channel_playlist_id[2:]
1766 return self.url_result(
1767 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1769 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1770 autogenerated = re.search(r'''(?x)
1772 channel-header-autogenerated-label|
1773 yt-channel-title-autogenerated
1774 )[^"]*"''', channel_page) is not None
1777 # The videos are contained in a single page
1778 # the ajax pages can't be used, they are empty
1781 video_id, 'Youtube', video_id=video_id,
1782 video_title=video_title)
1783 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1784 return self.playlist_result(entries, channel_id)
1786 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1789 class YoutubeUserIE(YoutubeChannelIE):
1790 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1791 _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_-]+)'
1792 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1793 IE_NAME = 'youtube:user'
1796 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1797 'playlist_mincount': 320,
1799 'title': 'TheLinuxFoundation',
1802 'url': 'ytuser:phihag',
1803 'only_matching': True,
1807 def suitable(cls, url):
1808 # Don't return True if the url can be extracted with other youtube
1809 # extractor, the regex would is too permissive and it would match.
1810 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1811 if any(ie.suitable(url) for ie in other_ies):
1814 return super(YoutubeUserIE, cls).suitable(url)
1817 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
1818 IE_DESC = 'YouTube.com user/channel playlists'
1819 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
1820 IE_NAME = 'youtube:playlists'
1823 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1824 'playlist_mincount': 4,
1826 'id': 'ThirstForScience',
1827 'title': 'Thirst for Science',
1830 # with "Load more" button
1831 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1832 'playlist_mincount': 70,
1835 'title': 'Игорь Клейнер',
1838 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
1839 'playlist_mincount': 17,
1841 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
1842 'title': 'Chem Player',
1847 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
1848 IE_DESC = 'YouTube.com searches'
1849 # there doesn't appear to be a real limit, for example if you search for
1850 # 'python' you get more than 8.000.000 results
1851 _MAX_RESULTS = float('inf')
1852 IE_NAME = 'youtube:search'
1853 _SEARCH_KEY = 'ytsearch'
1854 _EXTRA_QUERY_ARGS = {}
1857 def _get_n_results(self, query, n):
1858 """Get a specified number of results for a query"""
1863 for pagenum in itertools.count(1):
1865 'search_query': query.encode('utf-8'),
1869 url_query.update(self._EXTRA_QUERY_ARGS)
1870 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse.urlencode(url_query)
1871 data = self._download_json(
1872 result_url, video_id='query "%s"' % query,
1873 note='Downloading page %s' % pagenum,
1874 errnote='Unable to download API page')
1875 html_content = data[1]['body']['content']
1877 if 'class="search-message' in html_content:
1878 raise ExtractorError(
1879 '[youtube] No video results', expected=True)
1881 new_videos = self._ids_to_results(orderedSet(re.findall(
1882 r'href="/watch\?v=(.{11})', html_content)))
1883 videos += new_videos
1884 if not new_videos or len(videos) > limit:
1889 return self.playlist_result(videos, query)
1892 class YoutubeSearchDateIE(YoutubeSearchIE):
1893 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
1894 _SEARCH_KEY = 'ytsearchdate'
1895 IE_DESC = 'YouTube.com searches, newest videos first'
1896 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
1899 class YoutubeSearchURLIE(InfoExtractor):
1900 IE_DESC = 'YouTube.com search URLs'
1901 IE_NAME = 'youtube:search_url'
1902 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1904 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1905 'playlist_mincount': 5,
1907 'title': 'youtube-dl test video',
1911 def _real_extract(self, url):
1912 mobj = re.match(self._VALID_URL, url)
1913 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
1915 webpage = self._download_webpage(url, query)
1916 result_code = self._search_regex(
1917 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
1919 part_codes = re.findall(
1920 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
1922 for part_code in part_codes:
1923 part_title = self._html_search_regex(
1924 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
1925 part_url_snippet = self._html_search_regex(
1926 r'(?s)href="([^"]+)"', part_code, 'item URL')
1927 part_url = compat_urlparse.urljoin(
1928 'https://www.youtube.com/', part_url_snippet)
1932 'title': part_title,
1936 '_type': 'playlist',
1942 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
1943 IE_DESC = 'YouTube.com (multi-season) shows'
1944 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1945 IE_NAME = 'youtube:show'
1947 'url': 'https://www.youtube.com/show/airdisasters',
1948 'playlist_mincount': 5,
1950 'id': 'airdisasters',
1951 'title': 'Air Disasters',
1955 def _real_extract(self, url):
1956 playlist_id = self._match_id(url)
1957 return super(YoutubeShowIE, self)._real_extract(
1958 'https://www.youtube.com/show/%s/playlists' % playlist_id)
1961 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
1963 Base class for feed extractors
1964 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1966 _LOGIN_REQUIRED = True
1970 return 'youtube:%s' % self._FEED_NAME
1972 def _real_initialize(self):
1975 def _real_extract(self, url):
1976 page = self._download_webpage(
1977 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
1979 # The extraction process is the same as for playlists, but the regex
1980 # for the video ids doesn't contain an index
1982 more_widget_html = content_html = page
1983 for page_num in itertools.count(1):
1984 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
1986 # 'recommended' feed has infinite 'load more' and each new portion spins
1987 # the same videos in (sometimes) slightly different order, so we'll check
1988 # for unicity and break when portion has no new videos
1989 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
1995 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
1999 more = self._download_json(
2000 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2001 'Downloading page #%s' % page_num,
2002 transform_source=uppercase_escape)
2003 content_html = more['content_html']
2004 more_widget_html = more['load_more_widget_html']
2006 return self.playlist_result(
2007 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2010 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2011 IE_NAME = 'youtube:watchlater'
2012 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2013 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
2015 _TESTS = [] # override PlaylistIE tests
2017 def _real_extract(self, url):
2018 return self._extract_playlist('WL')
2021 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2022 IE_NAME = 'youtube:favorites'
2023 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2024 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2025 _LOGIN_REQUIRED = True
2027 def _real_extract(self, url):
2028 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2029 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2030 return self.url_result(playlist_id, 'YoutubePlaylist')
2033 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2034 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2035 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2036 _FEED_NAME = 'recommended'
2037 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2040 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2041 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2042 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2043 _FEED_NAME = 'subscriptions'
2044 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2047 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2048 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2049 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2050 _FEED_NAME = 'history'
2051 _PLAYLIST_TITLE = 'Youtube History'
2054 class YoutubeTruncatedURLIE(InfoExtractor):
2055 IE_NAME = 'youtube:truncated_url'
2056 IE_DESC = False # Do not list
2057 _VALID_URL = r'''(?x)
2059 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2062 annotation_id=annotation_[^&]+|
2068 attribution_link\?a=[^&]+
2074 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2075 'only_matching': True,
2077 'url': 'http://www.youtube.com/watch?',
2078 'only_matching': True,
2080 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2081 'only_matching': True,
2083 'url': 'https://www.youtube.com/watch?feature=foo',
2084 'only_matching': True,
2086 'url': 'https://www.youtube.com/watch?hl=en-GB',
2087 'only_matching': True,
2089 'url': 'https://www.youtube.com/watch?t=2372',
2090 'only_matching': True,
2093 def _real_extract(self, url):
2094 raise ExtractorError(
2095 'Did you forget to quote the URL? Remember that & is a meta '
2096 'character in most shells, so you want to put the URL in quotes, '
2098 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2099 ' or simply youtube-dl BaW_jenozKc .',
2103 class YoutubeTruncatedIDIE(InfoExtractor):
2104 IE_NAME = 'youtube:truncated_id'
2105 IE_DESC = False # Do not list
2106 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2109 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2110 'only_matching': True,
2113 def _real_extract(self, url):
2114 video_id = self._match_id(url)
2115 raise ExtractorError(
2116 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),