3 from __future__ import unicode_literals
14 from .common import InfoExtractor, SearchInfoExtractor
15 from ..jsinterp import JSInterpreter
16 from ..swfinterp import SWFInterpreter
17 from ..compat import (
20 compat_urllib_parse_unquote,
21 compat_urllib_parse_unquote_plus,
22 compat_urllib_parse_urlencode,
23 compat_urllib_parse_urlparse,
32 get_element_by_attribute,
52 class YoutubeBaseInfoExtractor(InfoExtractor):
53 """Provide base functions for Youtube extractors"""
54 _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
55 _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
56 _NETRC_MACHINE = 'youtube'
57 # If True it will raise an error if no login info is provided
58 _LOGIN_REQUIRED = False
60 def _set_language(self):
62 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
63 # YouTube sets the expire time to about two months
64 expire_time=time.time() + 2 * 30 * 24 * 3600)
66 def _ids_to_results(self, ids):
68 self.url_result(vid_id, 'Youtube', video_id=vid_id)
73 Attempt to log in to YouTube.
74 True is returned if successful or skipped.
75 False is returned if login failed.
77 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
79 (username, password) = self._get_login_info()
80 # No authentication to be performed
82 if self._LOGIN_REQUIRED:
83 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
86 login_page = self._download_webpage(
87 self._LOGIN_URL, None,
88 note='Downloading login page',
89 errnote='unable to fetch login page', fatal=False)
90 if login_page is False:
93 galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
94 login_page, 'Login GALX parameter')
98 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
103 'PersistentCookie': 'yes',
105 'bgresponse': 'js_disabled',
106 'checkConnection': '',
107 'checkedDomains': 'youtube',
114 'service': 'youtube',
119 login_data = urlencode_postdata(login_form_strs)
121 req = sanitized_Request(self._LOGIN_URL, login_data)
122 login_results = self._download_webpage(
124 note='Logging in', errnote='unable to log in', fatal=False)
125 if login_results is False:
128 if re.search(r'id="errormsg_0_Passwd"', login_results) is not None:
129 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected=True)
132 # TODO add SMS and phone call support - these require making a request and then prompting the user
134 if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
135 tfa_code = self._get_tfa_info('2-step verification code')
138 self._downloader.report_warning(
139 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
140 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
143 tfa_code = remove_start(tfa_code, 'G-')
145 tfa_form_strs = self._form_hidden_inputs('challenge', login_results)
147 tfa_form_strs.update({
152 tfa_data = urlencode_postdata(tfa_form_strs)
154 tfa_req = sanitized_Request(self._TWOFACTOR_URL, tfa_data)
155 tfa_results = self._download_webpage(
157 note='Submitting TFA code', errnote='unable to submit tfa', fatal=False)
159 if tfa_results is False:
162 if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
163 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
165 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
166 self._downloader.report_warning('unable to log in - did the page structure change?')
168 if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
169 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
172 if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
173 self._downloader.report_warning('unable to log in: bad username or password')
177 def _real_initialize(self):
178 if self._downloader is None:
181 if not self._login():
185 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
186 # Extract entries from page with "Load more" button
187 def _entries(self, page, playlist_id):
188 more_widget_html = content_html = page
189 for page_num in itertools.count(1):
190 for entry in self._process_page(content_html):
193 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
197 more = self._download_json(
198 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
199 'Downloading page #%s' % page_num,
200 transform_source=uppercase_escape)
201 content_html = more['content_html']
202 if not content_html.strip():
203 # Some webpages show a "Load more" button but they don't
206 more_widget_html = more['load_more_widget_html']
209 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
210 def _process_page(self, content):
211 for video_id, video_title in self.extract_videos_from_page(content):
212 yield self.url_result(video_id, 'Youtube', video_id, video_title)
214 def extract_videos_from_page(self, page):
217 for mobj in re.finditer(self._VIDEO_RE, page):
218 # The link with index 0 is not the first video of the playlist (not sure if still actual)
219 if 'index' in mobj.groupdict() and mobj.group('id') == '0':
221 video_id = mobj.group('id')
222 video_title = unescapeHTML(mobj.group('title'))
224 video_title = video_title.strip()
226 idx = ids_in_page.index(video_id)
227 if video_title and not titles_in_page[idx]:
228 titles_in_page[idx] = video_title
230 ids_in_page.append(video_id)
231 titles_in_page.append(video_title)
232 return zip(ids_in_page, titles_in_page)
235 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
236 def _process_page(self, content):
237 for playlist_id in orderedSet(re.findall(
238 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
240 yield self.url_result(
241 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
243 def _real_extract(self, url):
244 playlist_id = self._match_id(url)
245 webpage = self._download_webpage(url, playlist_id)
246 title = self._og_search_title(webpage, fatal=False)
247 return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
250 class YoutubeIE(YoutubeBaseInfoExtractor):
251 IE_DESC = 'YouTube.com'
252 _VALID_URL = r"""(?x)^
254 (?:https?://|//) # http(s):// or protocol-independent URL
255 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
256 (?:www\.)?deturl\.com/www\.youtube\.com/|
257 (?:www\.)?pwnyoutube\.com/|
258 (?:www\.)?yourepeat\.com/|
259 tube\.majestyc\.net/|
260 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
261 (?:.*?\#/)? # handle anchor (#/) redirect urls
262 (?: # the various things that can precede the ID:
263 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
264 |(?: # or the v= param in all its forms
265 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
266 (?:\?|\#!?) # the params delimiter ? or # or #!
267 (?:.*?[&;])?? # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&v=V36LpHqtcDY)
272 youtu\.be| # just youtu.be/xxxx
273 vid\.plus| # or vid.plus/xxxx
274 zwearz\.com/watch| # or zwearz.com/watch/xxxx
276 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
278 )? # all until now is optional -> you can pass the naked ID
279 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
280 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
281 (?(1).+)? # if we found the ID, everything can follow
283 _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
285 '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
286 '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
287 '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
288 '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
289 '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
290 '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
291 '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
292 '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
293 # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
294 '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
295 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
296 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
297 '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
298 '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
299 '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
300 '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
301 '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
302 '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
306 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
307 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
308 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
309 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
310 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
311 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
312 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
314 # Apple HTTP Live Streaming
315 '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
316 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
317 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
318 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
319 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
320 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
321 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
322 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
325 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
326 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
327 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
328 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
329 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
330 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
331 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
332 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
333 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
334 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60, 'preference': -40},
335 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264', 'preference': -40},
338 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
339 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
340 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
343 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
344 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
345 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
346 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
347 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
348 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40},
349 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9', 'preference': -40},
350 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
351 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
352 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
353 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
354 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
355 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
356 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
357 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
358 # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
359 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
360 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
361 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
362 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
363 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'preference': -40},
364 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60, 'preference': -40},
367 '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
368 '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
370 # Dash webm audio with opus inside
371 '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
372 '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
373 '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
376 '_rtmp': {'protocol': 'rtmp'},
378 _SUBTITLE_FORMATS = ('ttml', 'vtt')
383 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
387 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
388 'uploader': 'Philipp Hagemeister',
389 'uploader_id': 'phihag',
390 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
391 'upload_date': '20121002',
392 'license': 'Standard YouTube License',
393 '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 .',
394 'categories': ['Science & Technology'],
395 'tags': ['youtube-dl'],
397 'dislike_count': int,
403 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
404 'note': 'Test generic use_cipher_signature video (#897)',
408 'upload_date': '20120506',
409 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
410 'alt_title': 'I Love It (feat. Charli XCX)',
411 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
412 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
413 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
414 'iconic ep', 'iconic', 'love', 'it'],
415 'uploader': 'Icona Pop',
416 'uploader_id': 'IconaPop',
417 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IconaPop',
418 'license': 'Standard YouTube License',
419 'creator': 'Icona Pop',
423 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
424 'note': 'Test VEVO video with age protection (#956)',
428 'upload_date': '20130703',
429 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
430 'alt_title': 'Tunnel Vision',
431 'description': 'md5:64249768eec3bc4276236606ea996373',
432 'uploader': 'justintimberlakeVEVO',
433 'uploader_id': 'justintimberlakeVEVO',
434 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
435 'license': 'Standard YouTube License',
436 'creator': 'Justin Timberlake',
441 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
442 'note': 'Embed-only video (#1746)',
446 'upload_date': '20120608',
447 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
448 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
449 'uploader': 'SET India',
450 'uploader_id': 'setindia',
451 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/setindia',
452 'license': 'Standard YouTube License',
457 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
458 'note': 'Use the first video ID in the URL',
462 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
463 'uploader': 'Philipp Hagemeister',
464 'uploader_id': 'phihag',
465 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/phihag',
466 'upload_date': '20121002',
467 'license': 'Standard YouTube License',
468 '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 .',
469 'categories': ['Science & Technology'],
470 'tags': ['youtube-dl'],
472 'dislike_count': int,
475 'skip_download': True,
479 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
480 'note': '256k DASH audio (format 141) via DASH manifest',
484 'upload_date': '20121002',
485 'uploader_id': '8KVIDEO',
486 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
488 'uploader': '8KVIDEO',
489 'license': 'Standard YouTube License',
490 'title': 'UHDTV TEST 8K VIDEO.mp4'
493 'youtube_include_dash_manifest': True,
497 # DASH manifest with encrypted signature
499 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
503 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
504 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
505 'uploader': 'AfrojackVEVO',
506 'uploader_id': 'AfrojackVEVO',
507 'upload_date': '20131011',
508 'license': 'Standard YouTube License',
511 'youtube_include_dash_manifest': True,
515 # JS player signature function name containing $
517 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
521 'title': 'Taylor Swift - Shake It Off',
522 'alt_title': 'Shake It Off',
523 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
524 'uploader': 'TaylorSwiftVEVO',
525 'uploader_id': 'TaylorSwiftVEVO',
526 'upload_date': '20140818',
527 'license': 'Standard YouTube License',
528 'creator': 'Taylor Swift',
531 'youtube_include_dash_manifest': True,
537 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
541 'upload_date': '20100909',
542 'uploader': 'The Amazing Atheist',
543 'uploader_id': 'TheAmazingAtheist',
544 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
545 'license': 'Standard YouTube License',
546 'title': 'Burning Everyone\'s Koran',
547 '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',
550 # Normal age-gate video (No vevo, embed allowed)
552 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
556 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
557 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
558 'uploader': 'The Witcher',
559 'uploader_id': 'WitcherGame',
560 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
561 'upload_date': '20140605',
562 'license': 'Standard YouTube License',
566 # Age-gate video with encrypted signature
568 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
572 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
573 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
574 'uploader': 'LloydVEVO',
575 'uploader_id': 'LloydVEVO',
576 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
577 'upload_date': '20110629',
578 'license': 'Standard YouTube License',
582 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
584 'url': '__2ABJjxzNo',
588 'upload_date': '20100430',
589 'uploader_id': 'deadmau5',
590 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/deadmau5',
591 'creator': 'deadmau5',
592 'description': 'md5:12c56784b8032162bb936a5f76d55360',
593 'uploader': 'deadmau5',
594 'license': 'Standard YouTube License',
595 'title': 'Deadmau5 - Some Chords (HD)',
596 'alt_title': 'Some Chords',
598 'expected_warnings': [
599 'DASH manifest missing',
602 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
604 'url': 'lqQg6PlCWgI',
608 'upload_date': '20150827',
609 'uploader_id': 'olympic',
610 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/olympic',
611 'license': 'Standard YouTube License',
612 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
613 'uploader': 'Olympics',
614 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
617 'skip_download': 'requires avconv',
622 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
626 'stretched_ratio': 16 / 9.,
627 'upload_date': '20110310',
628 'uploader_id': 'AllenMeow',
629 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
630 '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',
632 'license': 'Standard YouTube License',
633 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
636 # url_encoded_fmt_stream_map is empty string
638 'url': 'qEJwOuvDf7I',
642 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
644 'upload_date': '20150404',
645 'uploader_id': 'spbelect',
646 'uploader': 'Наблюдатели Петербурга',
649 'skip_download': 'requires avconv',
651 'skip': 'This live event has ended.',
653 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
655 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
659 'title': 'md5:7b81415841e02ecd4313668cde88737a',
660 'description': 'md5:116377fd2963b81ec4ce64b542173306',
661 'upload_date': '20150625',
662 'uploader_id': 'dorappi2000',
663 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
664 'uploader': 'dorappi2000',
665 'license': 'Standard YouTube License',
666 'formats': 'mincount:33',
669 # DASH manifest with segment_list
671 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
672 'md5': '8ce563a1d667b599d21064e982ab9e31',
676 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
677 'uploader': 'Airtek',
678 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
679 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
680 'license': 'Standard YouTube License',
681 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
684 'youtube_include_dash_manifest': True,
685 'format': '135', # bestvideo
689 # Multifeed videos (multiple cameras), URL is for Main Camera
690 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
693 'title': 'teamPGP: Rocket League Noob Stream',
694 'description': 'md5:dc7872fb300e143831327f1bae3af010',
700 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
701 'description': 'md5:dc7872fb300e143831327f1bae3af010',
702 'upload_date': '20150721',
703 'uploader': 'Beer Games Beer',
704 'uploader_id': 'beergamesbeer',
705 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
706 'license': 'Standard YouTube License',
712 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
713 'description': 'md5:dc7872fb300e143831327f1bae3af010',
714 'upload_date': '20150721',
715 'uploader': 'Beer Games Beer',
716 'uploader_id': 'beergamesbeer',
717 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
718 'license': 'Standard YouTube License',
724 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
725 'description': 'md5:dc7872fb300e143831327f1bae3af010',
726 'upload_date': '20150721',
727 'uploader': 'Beer Games Beer',
728 'uploader_id': 'beergamesbeer',
729 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
730 'license': 'Standard YouTube License',
736 'title': 'teamPGP: Rocket League Noob Stream (zim)',
737 'description': 'md5:dc7872fb300e143831327f1bae3af010',
738 'upload_date': '20150721',
739 'uploader': 'Beer Games Beer',
740 'uploader_id': 'beergamesbeer',
741 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
742 'license': 'Standard YouTube License',
746 'skip_download': True,
750 # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
751 'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
754 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
759 'url': 'http://vid.plus/FlRa-iH7PGw',
760 'only_matching': True,
763 'url': 'http://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
764 'only_matching': True,
767 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
768 # Also tests cut-off URL expansion in video description (see
769 # https://github.com/rg3/youtube-dl/issues/1892,
770 # https://github.com/rg3/youtube-dl/issues/8164)
771 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
775 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
776 'alt_title': 'Dark Walk',
777 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
778 'upload_date': '20151119',
779 'uploader_id': 'IronSoulElf',
780 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
781 'uploader': 'IronSoulElf',
782 'license': 'Standard YouTube License',
783 'creator': 'Todd Haberman, Daniel Law Heath & Aaron Kaplan',
786 'skip_download': True,
790 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
791 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
792 'only_matching': True,
795 # Video with yt:stretch=17:0
796 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
800 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
801 'description': 'md5:ee18a25c350637c8faff806845bddee9',
802 'upload_date': '20151107',
803 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
804 'uploader': 'CH GAMER DROID',
807 'skip_download': True,
811 # Video licensed under Creative Commons
812 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
816 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
817 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
818 'upload_date': '20150127',
819 'uploader_id': 'BerkmanCenter',
820 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
821 'uploader': 'BerkmanCenter',
822 'license': 'Creative Commons Attribution license (reuse allowed)',
825 'skip_download': True,
829 # Channel-like uploader_url
830 'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
834 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
835 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
836 'upload_date': '20151119',
837 'uploader': 'Bernie 2016',
838 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
839 'uploader_url': 're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
840 'license': 'Creative Commons Attribution license (reuse allowed)',
843 'skip_download': True,
847 'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;v=V36LpHqtcDY',
848 'only_matching': True,
852 def __init__(self, *args, **kwargs):
853 super(YoutubeIE, self).__init__(*args, **kwargs)
854 self._player_cache = {}
856 def report_video_info_webpage_download(self, video_id):
857 """Report attempt to download video info webpage."""
858 self.to_screen('%s: Downloading video info webpage' % video_id)
860 def report_information_extraction(self, video_id):
861 """Report attempt to extract video information."""
862 self.to_screen('%s: Extracting video information' % video_id)
864 def report_unavailable_format(self, video_id, format):
865 """Report extracted video URL."""
866 self.to_screen('%s: Format %s not available' % (video_id, format))
868 def report_rtmp_download(self):
869 """Indicate the download will use the RTMP protocol."""
870 self.to_screen('RTMP download detected')
872 def _signature_cache_id(self, example_sig):
873 """ Return a string representation of a signature """
874 return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
876 def _extract_signature_function(self, video_id, player_url, example_sig):
878 r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
881 raise ExtractorError('Cannot identify player %r' % player_url)
882 player_type = id_m.group('ext')
883 player_id = id_m.group('id')
885 # Read from filesystem cache
886 func_id = '%s_%s_%s' % (
887 player_type, player_id, self._signature_cache_id(example_sig))
888 assert os.path.basename(func_id) == func_id
890 cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
891 if cache_spec is not None:
892 return lambda s: ''.join(s[i] for i in cache_spec)
895 'Downloading player %s' % player_url
896 if self._downloader.params.get('verbose') else
897 'Downloading %s player %s' % (player_type, player_id)
899 if player_type == 'js':
900 code = self._download_webpage(
901 player_url, video_id,
903 errnote='Download of %s failed' % player_url)
904 res = self._parse_sig_js(code)
905 elif player_type == 'swf':
906 urlh = self._request_webpage(
907 player_url, video_id,
909 errnote='Download of %s failed' % player_url)
911 res = self._parse_sig_swf(code)
913 assert False, 'Invalid player type %r' % player_type
915 test_string = ''.join(map(compat_chr, range(len(example_sig))))
916 cache_res = res(test_string)
917 cache_spec = [ord(c) for c in cache_res]
919 self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
922 def _print_sig_code(self, func, example_sig):
923 def gen_sig_code(idxs):
924 def _genslice(start, end, step):
925 starts = '' if start == 0 else str(start)
926 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
927 steps = '' if step == 1 else (':%d' % step)
928 return 's[%s%s%s]' % (starts, ends, steps)
931 # Quelch pyflakes warnings - start will be set when step is set
932 start = '(Never used)'
933 for i, prev in zip(idxs[1:], idxs[:-1]):
937 yield _genslice(start, prev, step)
940 if i - prev in [-1, 1]:
949 yield _genslice(start, i, step)
951 test_string = ''.join(map(compat_chr, range(len(example_sig))))
952 cache_res = func(test_string)
953 cache_spec = [ord(c) for c in cache_res]
954 expr_code = ' + '.join(gen_sig_code(cache_spec))
955 signature_id_tuple = '(%s)' % (
956 ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
957 code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
958 ' return %s\n') % (signature_id_tuple, expr_code)
959 self.to_screen('Extracted signature function:\n' + code)
961 def _parse_sig_js(self, jscode):
962 funcname = self._search_regex(
963 r'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode,
964 'Initial JS player signature function name')
966 jsi = JSInterpreter(jscode)
967 initial_function = jsi.extract_function(funcname)
968 return lambda s: initial_function([s])
970 def _parse_sig_swf(self, file_contents):
971 swfi = SWFInterpreter(file_contents)
972 TARGET_CLASSNAME = 'SignatureDecipher'
973 searched_class = swfi.extract_class(TARGET_CLASSNAME)
974 initial_function = swfi.extract_function(searched_class, 'decipher')
975 return lambda s: initial_function([s])
977 def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
978 """Turn the encrypted s field into a working signature"""
980 if player_url is None:
981 raise ExtractorError('Cannot decrypt signature without player_url')
983 if player_url.startswith('//'):
984 player_url = 'https:' + player_url
986 player_id = (player_url, self._signature_cache_id(s))
987 if player_id not in self._player_cache:
988 func = self._extract_signature_function(
989 video_id, player_url, s
991 self._player_cache[player_id] = func
992 func = self._player_cache[player_id]
993 if self._downloader.params.get('youtube_print_sig_code'):
994 self._print_sig_code(func, s)
996 except Exception as e:
997 tb = traceback.format_exc()
998 raise ExtractorError(
999 'Signature extraction failed: ' + tb, cause=e)
1001 def _get_subtitles(self, video_id, webpage):
1003 subs_doc = self._download_xml(
1004 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1005 video_id, note=False)
1006 except ExtractorError as err:
1007 self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1011 for track in subs_doc.findall('track'):
1012 lang = track.attrib['lang_code']
1013 if lang in sub_lang_list:
1016 for ext in self._SUBTITLE_FORMATS:
1017 params = compat_urllib_parse_urlencode({
1021 'name': track.attrib['name'].encode('utf-8'),
1023 sub_formats.append({
1024 'url': 'https://www.youtube.com/api/timedtext?' + params,
1027 sub_lang_list[lang] = sub_formats
1028 if not sub_lang_list:
1029 self._downloader.report_warning('video doesn\'t have subtitles')
1031 return sub_lang_list
1033 def _get_ytplayer_config(self, video_id, webpage):
1035 # User data may contain arbitrary character sequences that may affect
1036 # JSON extraction with regex, e.g. when '};' is contained the second
1037 # regex won't capture the whole JSON. Yet working around by trying more
1038 # concrete regex first keeping in mind proper quoted string handling
1039 # to be implemented in future that will replace this workaround (see
1040 # https://github.com/rg3/youtube-dl/issues/7468,
1041 # https://github.com/rg3/youtube-dl/pull/7599)
1042 r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1043 r';ytplayer\.config\s*=\s*({.+?});',
1045 config = self._search_regex(
1046 patterns, webpage, 'ytplayer.config', default=None)
1048 return self._parse_json(
1049 uppercase_escape(config), video_id, fatal=False)
1051 def _get_automatic_captions(self, video_id, webpage):
1052 """We need the webpage for getting the captions url, pass it as an
1053 argument to speed up the process."""
1054 self.to_screen('%s: Looking for automatic captions' % video_id)
1055 player_config = self._get_ytplayer_config(video_id, webpage)
1056 err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1057 if not player_config:
1058 self._downloader.report_warning(err_msg)
1061 args = player_config['args']
1062 caption_url = args.get('ttsurl')
1064 timestamp = args['timestamp']
1065 # We get the available subtitles
1066 list_params = compat_urllib_parse_urlencode({
1071 list_url = caption_url + '&' + list_params
1072 caption_list = self._download_xml(list_url, video_id)
1073 original_lang_node = caption_list.find('track')
1074 if original_lang_node is None:
1075 self._downloader.report_warning('Video doesn\'t have automatic captions')
1077 original_lang = original_lang_node.attrib['lang_code']
1078 caption_kind = original_lang_node.attrib.get('kind', '')
1081 for lang_node in caption_list.findall('target'):
1082 sub_lang = lang_node.attrib['lang_code']
1084 for ext in self._SUBTITLE_FORMATS:
1085 params = compat_urllib_parse_urlencode({
1086 'lang': original_lang,
1090 'kind': caption_kind,
1092 sub_formats.append({
1093 'url': caption_url + '&' + params,
1096 sub_lang_list[sub_lang] = sub_formats
1097 return sub_lang_list
1099 # Some videos don't provide ttsurl but rather caption_tracks and
1100 # caption_translation_languages (e.g. 20LmZk1hakA)
1101 caption_tracks = args['caption_tracks']
1102 caption_translation_languages = args['caption_translation_languages']
1103 caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1104 parsed_caption_url = compat_urllib_parse_urlparse(caption_url)
1105 caption_qs = compat_parse_qs(parsed_caption_url.query)
1108 for lang in caption_translation_languages.split(','):
1109 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1110 sub_lang = lang_qs.get('lc', [None])[0]
1114 for ext in self._SUBTITLE_FORMATS:
1116 'tlang': [sub_lang],
1119 sub_url = compat_urlparse.urlunparse(parsed_caption_url._replace(
1120 query=compat_urllib_parse_urlencode(caption_qs, True)))
1121 sub_formats.append({
1125 sub_lang_list[sub_lang] = sub_formats
1126 return sub_lang_list
1127 # An extractor error can be raise by the download process if there are
1128 # no automatic captions but there are subtitles
1129 except (KeyError, ExtractorError):
1130 self._downloader.report_warning(err_msg)
1133 def _mark_watched(self, video_id, video_info):
1134 playback_url = video_info.get('videostats_playback_base_url', [None])[0]
1135 if not playback_url:
1137 parsed_playback_url = compat_urlparse.urlparse(playback_url)
1138 qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1140 # cpn generation algorithm is reverse engineered from base.js.
1141 # In fact it works even with dummy cpn.
1142 CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1143 cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1149 playback_url = compat_urlparse.urlunparse(
1150 parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1152 self._download_webpage(
1153 playback_url, video_id, 'Marking watched',
1154 'Unable to mark watched', fatal=False)
1157 def extract_id(cls, url):
1158 mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1160 raise ExtractorError('Invalid URL: %s' % url)
1161 video_id = mobj.group(2)
1164 def _extract_from_m3u8(self, manifest_url, video_id):
1167 def _get_urls(_manifest):
1168 lines = _manifest.split('\n')
1169 urls = filter(lambda l: l and not l.startswith('#'),
1172 manifest = self._download_webpage(manifest_url, video_id, 'Downloading formats manifest')
1173 formats_urls = _get_urls(manifest)
1174 for format_url in formats_urls:
1175 itag = self._search_regex(r'itag/(\d+?)/', format_url, 'itag')
1176 url_map[itag] = format_url
1179 def _extract_annotations(self, video_id):
1180 url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1181 return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
1183 def _real_extract(self, url):
1184 url, smuggled_data = unsmuggle_url(url, {})
1187 'http' if self._downloader.params.get('prefer_insecure', False)
1192 parsed_url = compat_urllib_parse_urlparse(url)
1193 for component in [parsed_url.fragment, parsed_url.query]:
1194 query = compat_parse_qs(component)
1195 if start_time is None and 't' in query:
1196 start_time = parse_duration(query['t'][0])
1197 if start_time is None and 'start' in query:
1198 start_time = parse_duration(query['start'][0])
1199 if end_time is None and 'end' in query:
1200 end_time = parse_duration(query['end'][0])
1202 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1203 mobj = re.search(self._NEXT_URL_RE, url)
1205 url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1206 video_id = self.extract_id(url)
1209 url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1210 video_webpage = self._download_webpage(url, video_id)
1212 # Attempt to extract SWF player URL
1213 mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1214 if mobj is not None:
1215 player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1221 def add_dash_mpd(video_info):
1222 dash_mpd = video_info.get('dashmpd')
1223 if dash_mpd and dash_mpd[0] not in dash_mpds:
1224 dash_mpds.append(dash_mpd[0])
1227 embed_webpage = None
1229 if re.search(r'player-age-gate-content">', video_webpage) is not None:
1231 # We simulate the access to the video from www.youtube.com/v/{video_id}
1232 # this can be viewed without login into Youtube
1233 url = proto + '://www.youtube.com/embed/%s' % video_id
1234 embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1235 data = compat_urllib_parse_urlencode({
1236 'video_id': video_id,
1237 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1238 'sts': self._search_regex(
1239 r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1241 video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1242 video_info_webpage = self._download_webpage(
1243 video_info_url, video_id,
1244 note='Refetching age-gated info webpage',
1245 errnote='unable to download video info webpage')
1246 video_info = compat_parse_qs(video_info_webpage)
1247 add_dash_mpd(video_info)
1251 # Try looking directly into the video webpage
1252 ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1254 args = ytplayer_config['args']
1255 if args.get('url_encoded_fmt_stream_map'):
1256 # Convert to the same format returned by compat_parse_qs
1257 video_info = dict((k, [v]) for k, v in args.items())
1258 add_dash_mpd(video_info)
1259 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1261 if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1262 # We also try looking in get_video_info since it may contain different dashmpd
1263 # URL that points to a DASH manifest with possibly different itag set (some itags
1264 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1265 # manifest pointed by get_video_info's dashmpd).
1266 # The general idea is to take a union of itags of both DASH manifests (for example
1267 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1268 self.report_video_info_webpage_download(video_id)
1269 for el_type in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1271 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1272 % (proto, video_id, el_type))
1273 video_info_webpage = self._download_webpage(
1275 video_id, note=False,
1276 errnote='unable to download video info webpage')
1277 get_video_info = compat_parse_qs(video_info_webpage)
1278 if get_video_info.get('use_cipher_signature') != ['True']:
1279 add_dash_mpd(get_video_info)
1281 video_info = get_video_info
1282 if 'token' in get_video_info:
1283 # Different get_video_info requests may report different results, e.g.
1284 # some may report video unavailability, but some may serve it without
1285 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1286 # the original webpage as well as el=info and el=embedded get_video_info
1287 # requests report video unavailability due to geo restriction while
1288 # el=detailpage succeeds and returns valid data). This is probably
1289 # due to YouTube measures against IP ranges of hosting providers.
1290 # Working around by preferring the first succeeded video_info containing
1291 # the token if no such video_info yet was found.
1292 if 'token' not in video_info:
1293 video_info = get_video_info
1295 if 'token' not in video_info:
1296 if 'reason' in video_info:
1297 if 'The uploader has not made this video available in your country.' in video_info['reason']:
1298 regions_allowed = self._html_search_meta('regionsAllowed', video_webpage, default=None)
1300 raise ExtractorError('YouTube said: This video is available in %s only' % (
1301 ', '.join(map(ISO3166Utils.short2full, regions_allowed.split(',')))),
1303 raise ExtractorError(
1304 'YouTube said: %s' % video_info['reason'][0],
1305 expected=True, video_id=video_id)
1307 raise ExtractorError(
1308 '"token" parameter not in video info for unknown reason',
1312 if 'title' in video_info:
1313 video_title = video_info['title'][0]
1315 self._downloader.report_warning('Unable to extract video title')
1319 video_description = get_element_by_id("eow-description", video_webpage)
1320 if video_description:
1321 video_description = re.sub(r'''(?x)
1323 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1324 (?:title|href)="([^"]+)"\s+
1325 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1326 class="(?:yt-uix-redirect-link|yt-uix-sessionlink[^"]*)"[^>]*>
1329 ''', r'\1', video_description)
1330 video_description = clean_html(video_description)
1332 fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1334 video_description = unescapeHTML(fd_mobj.group(1))
1336 video_description = ''
1338 if 'multifeed_metadata_list' in video_info and not smuggled_data.get('force_singlefeed', False):
1339 if not self._downloader.params.get('noplaylist'):
1342 multifeed_metadata_list = video_info['multifeed_metadata_list'][0]
1343 for feed in multifeed_metadata_list.split(','):
1344 # Unquote should take place before split on comma (,) since textual
1345 # fields may contain comma as well (see
1346 # https://github.com/rg3/youtube-dl/issues/8536)
1347 feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1349 '_type': 'url_transparent',
1350 'ie_key': 'Youtube',
1352 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1353 {'force_singlefeed': True}),
1354 'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1356 feed_ids.append(feed_data['id'][0])
1358 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1359 % (', '.join(feed_ids), video_id))
1360 return self.playlist_result(entries, video_id, video_title, video_description)
1361 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1363 if 'view_count' in video_info:
1364 view_count = int(video_info['view_count'][0])
1368 # Check for "rental" videos
1369 if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1370 raise ExtractorError('"rental" videos not supported')
1372 # Start extracting information
1373 self.report_information_extraction(video_id)
1376 if 'author' not in video_info:
1377 raise ExtractorError('Unable to extract uploader name')
1378 video_uploader = compat_urllib_parse_unquote_plus(video_info['author'][0])
1381 video_uploader_id = None
1382 video_uploader_url = None
1384 r'<link itemprop="url" href="(?P<uploader_url>https?://www.youtube.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
1386 if mobj is not None:
1387 video_uploader_id = mobj.group('uploader_id')
1388 video_uploader_url = mobj.group('uploader_url')
1390 self._downloader.report_warning('unable to extract uploader nickname')
1393 # We try first to get a high quality image:
1394 m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
1395 video_webpage, re.DOTALL)
1396 if m_thumb is not None:
1397 video_thumbnail = m_thumb.group(1)
1398 elif 'thumbnail_url' not in video_info:
1399 self._downloader.report_warning('unable to extract video thumbnail')
1400 video_thumbnail = None
1401 else: # don't panic if we can't find it
1402 video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
1405 upload_date = self._html_search_meta(
1406 'datePublished', video_webpage, 'upload date', default=None)
1408 upload_date = self._search_regex(
1409 [r'(?s)id="eow-date.*?>(.*?)</span>',
1410 r'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1411 video_webpage, 'upload date', default=None)
1413 upload_date = ' '.join(re.sub(r'[/,-]', r' ', mobj.group(1)).split())
1414 upload_date = unified_strdate(upload_date)
1416 video_license = self._html_search_regex(
1417 r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
1418 video_webpage, 'license', default=None)
1420 m_music = re.search(
1421 r'<h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*<ul[^>]*>\s*<li>(?P<title>.+?) by (?P<creator>.+?)(?:\(.+?\))?</li',
1424 video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
1425 video_creator = clean_html(m_music.group('creator'))
1427 video_alt_title = video_creator = None
1429 m_cat_container = self._search_regex(
1430 r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1431 video_webpage, 'categories', default=None)
1433 category = self._html_search_regex(
1434 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
1436 video_categories = None if category is None else [category]
1438 video_categories = None
1441 unescapeHTML(m.group('content'))
1442 for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
1444 def _extract_count(count_name):
1445 return str_to_int(self._search_regex(
1446 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1447 % re.escape(count_name),
1448 video_webpage, count_name, default=None))
1450 like_count = _extract_count('like')
1451 dislike_count = _extract_count('dislike')
1454 video_subtitles = self.extract_subtitles(video_id, video_webpage)
1455 automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
1457 if 'length_seconds' not in video_info:
1458 self._downloader.report_warning('unable to extract video duration')
1459 video_duration = None
1461 video_duration = int(compat_urllib_parse_unquote_plus(video_info['length_seconds'][0]))
1464 video_annotations = None
1465 if self._downloader.params.get('writeannotations', False):
1466 video_annotations = self._extract_annotations(video_id)
1468 def _map_to_format_list(urlmap):
1470 for itag, video_real_url in urlmap.items():
1473 'url': video_real_url,
1474 'player_url': player_url,
1476 if itag in self._formats:
1477 dct.update(self._formats[itag])
1481 if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1482 self.report_rtmp_download()
1484 'format_id': '_rtmp',
1486 'url': video_info['conn'][0],
1487 'player_url': player_url,
1489 elif len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1:
1490 encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1491 if 'rtmpe%3Dyes' in encoded_url_map:
1492 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
1494 fmt_list = video_info.get('fmt_list', [''])[0]
1496 for fmt in fmt_list.split(','):
1497 spec = fmt.split('/')
1499 width_height = spec[1].split('x')
1500 if len(width_height) == 2:
1501 formats_spec[spec[0]] = {
1502 'resolution': spec[1],
1503 'width': int_or_none(width_height[0]),
1504 'height': int_or_none(width_height[1]),
1507 for url_data_str in encoded_url_map.split(','):
1508 url_data = compat_parse_qs(url_data_str)
1509 if 'itag' not in url_data or 'url' not in url_data:
1511 format_id = url_data['itag'][0]
1512 url = url_data['url'][0]
1514 if 'sig' in url_data:
1515 url += '&signature=' + url_data['sig'][0]
1516 elif 's' in url_data:
1517 encrypted_sig = url_data['s'][0]
1518 ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1520 jsplayer_url_json = self._search_regex(
1522 embed_webpage if age_gate else video_webpage,
1523 'JS player URL (1)', default=None)
1524 if not jsplayer_url_json and not age_gate:
1525 # We need the embed website after all
1526 if embed_webpage is None:
1527 embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1528 embed_webpage = self._download_webpage(
1529 embed_url, video_id, 'Downloading embed webpage')
1530 jsplayer_url_json = self._search_regex(
1531 ASSETS_RE, embed_webpage, 'JS player URL')
1533 player_url = json.loads(jsplayer_url_json)
1534 if player_url is None:
1535 player_url_json = self._search_regex(
1536 r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1537 video_webpage, 'age gate player URL')
1538 player_url = json.loads(player_url_json)
1540 if self._downloader.params.get('verbose'):
1541 if player_url is None:
1542 player_version = 'unknown'
1543 player_desc = 'unknown'
1545 if player_url.endswith('swf'):
1546 player_version = self._search_regex(
1547 r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1548 'flash player', fatal=False)
1549 player_desc = 'flash player %s' % player_version
1551 player_version = self._search_regex(
1552 [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r'(?:www|player)-([^/]+)/base\.js'],
1554 'html5 player', fatal=False)
1555 player_desc = 'html5 player %s' % player_version
1557 parts_sizes = self._signature_cache_id(encrypted_sig)
1558 self.to_screen('{%s} signature length %s, %s' %
1559 (format_id, parts_sizes, player_desc))
1561 signature = self._decrypt_signature(
1562 encrypted_sig, video_id, player_url, age_gate)
1563 url += '&signature=' + signature
1564 if 'ratebypass' not in url:
1565 url += '&ratebypass=yes'
1568 'format_id': format_id,
1570 'player_url': player_url,
1572 if format_id in self._formats:
1573 dct.update(self._formats[format_id])
1574 if format_id in formats_spec:
1575 dct.update(formats_spec[format_id])
1577 # Some itags are not included in DASH manifest thus corresponding formats will
1578 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1579 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1580 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
1581 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
1584 'filesize': int_or_none(url_data.get('clen', [None])[0]),
1585 'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
1588 'fps': int_or_none(url_data.get('fps', [None])[0]),
1589 'format_note': url_data.get('quality_label', [None])[0] or url_data.get('quality', [None])[0],
1591 for key, value in more_fields.items():
1594 type_ = url_data.get('type', [None])[0]
1596 type_split = type_.split(';')
1597 kind_ext = type_split[0].split('/')
1598 if len(kind_ext) == 2:
1600 dct['ext'] = mimetype2ext(type_split[0])
1601 if kind in ('audio', 'video'):
1603 for mobj in re.finditer(
1604 r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
1605 if mobj.group('key') == 'codecs':
1606 codecs = mobj.group('val')
1609 codecs = codecs.split(',')
1610 if len(codecs) == 2:
1611 acodec, vcodec = codecs[1], codecs[0]
1613 acodec, vcodec = (codecs[0], 'none') if kind == 'audio' else ('none', codecs[0])
1619 elif video_info.get('hlsvp'):
1620 manifest_url = video_info['hlsvp'][0]
1621 url_map = self._extract_from_m3u8(manifest_url, video_id)
1622 formats = _map_to_format_list(url_map)
1623 # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
1624 for a_format in formats:
1625 a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
1627 unavailable_message = self._html_search_regex(
1628 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1629 video_webpage, 'unavailable message', default=None)
1630 if unavailable_message:
1631 raise ExtractorError(unavailable_message, expected=True)
1632 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1634 # Look for the DASH manifest
1635 if self._downloader.params.get('youtube_include_dash_manifest', True):
1636 dash_mpd_fatal = True
1637 for mpd_url in dash_mpds:
1640 def decrypt_sig(mobj):
1642 dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
1643 return '/signature/%s' % dec_s
1645 mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
1647 for df in self._extract_mpd_formats(
1648 mpd_url, video_id, fatal=dash_mpd_fatal,
1649 formats_dict=self._formats):
1650 # Do not overwrite DASH format found in some previous DASH manifest
1651 if df['format_id'] not in dash_formats:
1652 dash_formats[df['format_id']] = df
1653 # Additional DASH manifests may end up in HTTP Error 403 therefore
1654 # allow them to fail without bug report message if we already have
1655 # some DASH manifest succeeded. This is temporary workaround to reduce
1656 # burst of bug reports until we figure out the reason and whether it
1657 # can be fixed at all.
1658 dash_mpd_fatal = False
1659 except (ExtractorError, KeyError) as e:
1660 self.report_warning(
1661 'Skipping DASH manifest: %r' % e, video_id)
1663 # Remove the formats we found through non-DASH, they
1664 # contain less info and it can be wrong, because we use
1665 # fixed values (for example the resolution). See
1666 # https://github.com/rg3/youtube-dl/issues/5774 for an
1668 formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
1669 formats.extend(dash_formats.values())
1671 # Check for malformed aspect ratio
1672 stretched_m = re.search(
1673 r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1676 w = float(stretched_m.group('w'))
1677 h = float(stretched_m.group('h'))
1678 # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
1679 # We will only process correct ratios.
1683 if f.get('vcodec') != 'none':
1684 f['stretched_ratio'] = ratio
1686 self._sort_formats(formats)
1688 self.mark_watched(video_id, video_info)
1692 'uploader': video_uploader,
1693 'uploader_id': video_uploader_id,
1694 'uploader_url': video_uploader_url,
1695 'upload_date': upload_date,
1696 'license': video_license,
1697 'creator': video_creator,
1698 'title': video_title,
1699 'alt_title': video_alt_title,
1700 'thumbnail': video_thumbnail,
1701 'description': video_description,
1702 'categories': video_categories,
1704 'subtitles': video_subtitles,
1705 'automatic_captions': automatic_captions,
1706 'duration': video_duration,
1707 'age_limit': 18 if age_gate else 0,
1708 'annotations': video_annotations,
1709 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
1710 'view_count': view_count,
1711 'like_count': like_count,
1712 'dislike_count': dislike_count,
1713 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
1716 'start_time': start_time,
1717 'end_time': end_time,
1721 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
1722 IE_DESC = 'YouTube.com playlists'
1723 _VALID_URL = r"""(?x)(?:
1728 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1729 \? (?:.*?[&;])*? (?:p|a|list)=
1733 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1734 # Top tracks, they can also include dots
1739 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1741 _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
1742 _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
1743 IE_NAME = 'youtube:playlist'
1745 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1747 'title': 'ytdl test PL',
1748 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1750 'playlist_count': 3,
1752 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1754 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1755 'title': 'YDL_Empty_List',
1757 'playlist_count': 0,
1759 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1760 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1762 'title': '29C3: Not my department',
1763 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1765 'playlist_count': 95,
1767 'note': 'issue #673',
1768 'url': 'PLBB231211A4F62143',
1770 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1771 'id': 'PLBB231211A4F62143',
1773 'playlist_mincount': 26,
1775 'note': 'Large playlist',
1776 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1778 'title': 'Uploads from Cauchemar',
1779 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1781 'playlist_mincount': 799,
1783 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1785 'title': 'YDL_safe_search',
1786 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1788 'playlist_count': 2,
1791 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1792 'playlist_count': 4,
1795 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1798 'note': 'Embedded SWF player',
1799 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1800 'playlist_count': 4,
1803 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1806 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1807 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1809 'title': 'Uploads from Interstellar Movie',
1810 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1812 'playlist_mincout': 21,
1815 def _real_initialize(self):
1818 def _extract_mix(self, playlist_id):
1819 # The mixes are generated from a single video
1820 # the id of the playlist is just 'RD' + video_id
1821 url = 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id[-11:], playlist_id)
1822 webpage = self._download_webpage(
1823 url, playlist_id, 'Downloading Youtube mix')
1824 search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
1826 search_title('playlist-title') or
1827 search_title('title long-title') or
1828 search_title('title'))
1829 title = clean_html(title_span)
1830 ids = orderedSet(re.findall(
1831 r'''(?xs)data-video-username=".*?".*?
1832 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re.escape(playlist_id),
1834 url_results = self._ids_to_results(ids)
1836 return self.playlist_result(url_results, playlist_id, title)
1838 def _extract_playlist(self, playlist_id):
1839 url = self._TEMPLATE_URL % playlist_id
1840 page = self._download_webpage(url, playlist_id)
1842 for match in re.findall(r'<div class="yt-alert-message">([^<]+)</div>', page):
1843 match = match.strip()
1844 # Check if the playlist exists or is private
1845 if re.match(r'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match):
1846 raise ExtractorError(
1847 'The playlist doesn\'t exist or is private, use --username or '
1848 '--netrc to access it.',
1850 elif re.match(r'[^<]*Invalid parameters[^<]*', match):
1851 raise ExtractorError(
1852 'Invalid parameters. Maybe URL is incorrect.',
1854 elif re.match(r'[^<]*Choose your language[^<]*', match):
1857 self.report_warning('Youtube gives an alert message: ' + match)
1859 playlist_title = self._html_search_regex(
1860 r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1863 return self.playlist_result(self._entries(page, playlist_id), playlist_id, playlist_title)
1865 def _check_download_just_video(self, url, playlist_id):
1866 # Check if it's a video-specific URL
1867 query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
1868 if 'v' in query_dict:
1869 video_id = query_dict['v'][0]
1870 if self._downloader.params.get('noplaylist'):
1871 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1872 return self.url_result(video_id, 'Youtube', video_id=video_id)
1874 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
1876 def _real_extract(self, url):
1877 # Extract playlist id
1878 mobj = re.match(self._VALID_URL, url)
1880 raise ExtractorError('Invalid URL: %s' % url)
1881 playlist_id = mobj.group(1) or mobj.group(2)
1883 video = self._check_download_just_video(url, playlist_id)
1887 if playlist_id.startswith(('RD', 'UL', 'PU')):
1888 # Mixes require a custom extraction process
1889 return self._extract_mix(playlist_id)
1891 return self._extract_playlist(playlist_id)
1894 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
1895 IE_DESC = 'YouTube.com channels'
1896 _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1897 _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
1898 _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1899 IE_NAME = 'youtube:channel'
1901 'note': 'paginated channel',
1902 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1903 'playlist_mincount': 91,
1905 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1906 'title': 'Uploads from lex will',
1909 'note': 'Age restricted channel',
1910 # from https://www.youtube.com/user/DeusExOfficial
1911 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1912 'playlist_mincount': 64,
1914 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1915 'title': 'Uploads from Deus Ex',
1920 def suitable(cls, url):
1921 return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
1922 else super(YoutubeChannelIE, cls).suitable(url))
1924 def _real_extract(self, url):
1925 channel_id = self._match_id(url)
1927 url = self._TEMPLATE_URL % channel_id
1929 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1930 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1931 # otherwise fallback on channel by page extraction
1932 channel_page = self._download_webpage(
1933 url + '?view=57', channel_id,
1934 'Downloading channel page', fatal=False)
1935 if channel_page is False:
1936 channel_playlist_id = False
1938 channel_playlist_id = self._html_search_meta(
1939 'channelId', channel_page, 'channel id', default=None)
1940 if not channel_playlist_id:
1941 channel_playlist_id = self._search_regex(
1942 r'data-(?:channel-external-|yt)id="([^"]+)"',
1943 channel_page, 'channel id', default=None)
1944 if channel_playlist_id and channel_playlist_id.startswith('UC'):
1945 playlist_id = 'UU' + channel_playlist_id[2:]
1946 return self.url_result(
1947 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
1949 channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
1950 autogenerated = re.search(r'''(?x)
1952 channel-header-autogenerated-label|
1953 yt-channel-title-autogenerated
1954 )[^"]*"''', channel_page) is not None
1957 # The videos are contained in a single page
1958 # the ajax pages can't be used, they are empty
1961 video_id, 'Youtube', video_id=video_id,
1962 video_title=video_title)
1963 for video_id, video_title in self.extract_videos_from_page(channel_page)]
1964 return self.playlist_result(entries, channel_id)
1966 return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
1969 class YoutubeUserIE(YoutubeChannelIE):
1970 IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
1971 _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_-]+)'
1972 _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
1973 IE_NAME = 'youtube:user'
1976 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1977 'playlist_mincount': 320,
1979 'title': 'TheLinuxFoundation',
1982 'url': 'ytuser:phihag',
1983 'only_matching': True,
1987 def suitable(cls, url):
1988 # Don't return True if the url can be extracted with other youtube
1989 # extractor, the regex would is too permissive and it would match.
1990 other_ies = iter(klass for (name, klass) in globals().items() if name.endswith('IE') and klass is not cls)
1991 if any(ie.suitable(url) for ie in other_ies):
1994 return super(YoutubeUserIE, cls).suitable(url)
1997 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
1998 IE_DESC = 'YouTube.com live streams'
1999 _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+))/live'
2000 IE_NAME = 'youtube:live'
2003 'url': 'http://www.youtube.com/user/TheYoungTurks/live',
2005 'id': 'a48o2S1cPoo',
2007 'title': 'The Young Turks - Live Main Show',
2008 'uploader': 'The Young Turks',
2009 'uploader_id': 'TheYoungTurks',
2010 'uploader_url': 're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2011 'upload_date': '20150715',
2012 'license': 'Standard YouTube License',
2013 'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2014 'categories': ['News & Politics'],
2015 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2017 'dislike_count': int,
2020 'skip_download': True,
2023 'url': 'http://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2024 'only_matching': True,
2027 def _real_extract(self, url):
2028 mobj = re.match(self._VALID_URL, url)
2029 channel_id = mobj.group('id')
2030 base_url = mobj.group('base_url')
2031 webpage = self._download_webpage(url, channel_id, fatal=False)
2033 page_type = self._og_search_property(
2034 'type', webpage, 'page type', default=None)
2035 video_id = self._html_search_meta(
2036 'videoId', webpage, 'video id', default=None)
2037 if page_type == 'video' and video_id and re.match(r'^[0-9A-Za-z_-]{11}$', video_id):
2038 return self.url_result(video_id, YoutubeIE.ie_key())
2039 return self.url_result(base_url)
2042 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2043 IE_DESC = 'YouTube.com user/channel playlists'
2044 _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2045 IE_NAME = 'youtube:playlists'
2048 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
2049 'playlist_mincount': 4,
2051 'id': 'ThirstForScience',
2052 'title': 'Thirst for Science',
2055 # with "Load more" button
2056 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2057 'playlist_mincount': 70,
2060 'title': 'Игорь Клейнер',
2063 'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2064 'playlist_mincount': 17,
2066 'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2067 'title': 'Chem Player',
2072 class YoutubeSearchIE(SearchInfoExtractor, YoutubePlaylistIE):
2073 IE_DESC = 'YouTube.com searches'
2074 # there doesn't appear to be a real limit, for example if you search for
2075 # 'python' you get more than 8.000.000 results
2076 _MAX_RESULTS = float('inf')
2077 IE_NAME = 'youtube:search'
2078 _SEARCH_KEY = 'ytsearch'
2079 _EXTRA_QUERY_ARGS = {}
2082 def _get_n_results(self, query, n):
2083 """Get a specified number of results for a query"""
2088 for pagenum in itertools.count(1):
2090 'search_query': query.encode('utf-8'),
2094 url_query.update(self._EXTRA_QUERY_ARGS)
2095 result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2096 data = self._download_json(
2097 result_url, video_id='query "%s"' % query,
2098 note='Downloading page %s' % pagenum,
2099 errnote='Unable to download API page')
2100 html_content = data[1]['body']['content']
2102 if 'class="search-message' in html_content:
2103 raise ExtractorError(
2104 '[youtube] No video results', expected=True)
2106 new_videos = self._ids_to_results(orderedSet(re.findall(
2107 r'href="/watch\?v=(.{11})', html_content)))
2108 videos += new_videos
2109 if not new_videos or len(videos) > limit:
2114 return self.playlist_result(videos, query)
2117 class YoutubeSearchDateIE(YoutubeSearchIE):
2118 IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2119 _SEARCH_KEY = 'ytsearchdate'
2120 IE_DESC = 'YouTube.com searches, newest videos first'
2121 _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2124 class YoutubeSearchURLIE(InfoExtractor):
2125 IE_DESC = 'YouTube.com search URLs'
2126 IE_NAME = 'youtube:search_url'
2127 _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2129 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2130 'playlist_mincount': 5,
2132 'title': 'youtube-dl test video',
2135 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
2136 'only_matching': True,
2139 def _real_extract(self, url):
2140 mobj = re.match(self._VALID_URL, url)
2141 query = compat_urllib_parse_unquote_plus(mobj.group('query'))
2143 webpage = self._download_webpage(url, query)
2144 result_code = self._search_regex(
2145 r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
2147 part_codes = re.findall(
2148 r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
2150 for part_code in part_codes:
2151 part_title = self._html_search_regex(
2152 [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
2153 part_url_snippet = self._html_search_regex(
2154 r'(?s)href="([^"]+)"', part_code, 'item URL')
2155 part_url = compat_urlparse.urljoin(
2156 'https://www.youtube.com/', part_url_snippet)
2160 'title': part_title,
2164 '_type': 'playlist',
2170 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
2171 IE_DESC = 'YouTube.com (multi-season) shows'
2172 _VALID_URL = r'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
2173 IE_NAME = 'youtube:show'
2175 'url': 'https://www.youtube.com/show/airdisasters',
2176 'playlist_mincount': 5,
2178 'id': 'airdisasters',
2179 'title': 'Air Disasters',
2183 def _real_extract(self, url):
2184 playlist_id = self._match_id(url)
2185 return super(YoutubeShowIE, self)._real_extract(
2186 'https://www.youtube.com/show/%s/playlists' % playlist_id)
2189 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
2191 Base class for feed extractors
2192 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
2194 _LOGIN_REQUIRED = True
2198 return 'youtube:%s' % self._FEED_NAME
2200 def _real_initialize(self):
2203 def _real_extract(self, url):
2204 page = self._download_webpage(
2205 'https://www.youtube.com/feed/%s' % self._FEED_NAME, self._PLAYLIST_TITLE)
2207 # The extraction process is the same as for playlists, but the regex
2208 # for the video ids doesn't contain an index
2210 more_widget_html = content_html = page
2211 for page_num in itertools.count(1):
2212 matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
2214 # 'recommended' feed has infinite 'load more' and each new portion spins
2215 # the same videos in (sometimes) slightly different order, so we'll check
2216 # for unicity and break when portion has no new videos
2217 new_ids = filter(lambda video_id: video_id not in ids, orderedSet(matches))
2223 mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
2227 more = self._download_json(
2228 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
2229 'Downloading page #%s' % page_num,
2230 transform_source=uppercase_escape)
2231 content_html = more['content_html']
2232 more_widget_html = more['load_more_widget_html']
2234 return self.playlist_result(
2235 self._ids_to_results(ids), playlist_title=self._PLAYLIST_TITLE)
2238 class YoutubeWatchLaterIE(YoutubePlaylistIE):
2239 IE_NAME = 'youtube:watchlater'
2240 IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2241 _VALID_URL = r'https?://www\.youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
2244 'url': 'https://www.youtube.com/playlist?list=WL',
2245 'only_matching': True,
2247 'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
2248 'only_matching': True,
2251 def _real_extract(self, url):
2252 video = self._check_download_just_video(url, 'WL')
2255 return self._extract_playlist('WL')
2258 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
2259 IE_NAME = 'youtube:favorites'
2260 IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2261 _VALID_URL = r'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2262 _LOGIN_REQUIRED = True
2264 def _real_extract(self, url):
2265 webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2266 playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
2267 return self.url_result(playlist_id, 'YoutubePlaylist')
2270 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
2271 IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2272 _VALID_URL = r'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2273 _FEED_NAME = 'recommended'
2274 _PLAYLIST_TITLE = 'Youtube Recommended videos'
2277 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
2278 IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2279 _VALID_URL = r'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2280 _FEED_NAME = 'subscriptions'
2281 _PLAYLIST_TITLE = 'Youtube Subscriptions'
2284 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
2285 IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
2286 _VALID_URL = 'https?://www\.youtube\.com/feed/history|:ythistory'
2287 _FEED_NAME = 'history'
2288 _PLAYLIST_TITLE = 'Youtube History'
2291 class YoutubeTruncatedURLIE(InfoExtractor):
2292 IE_NAME = 'youtube:truncated_url'
2293 IE_DESC = False # Do not list
2294 _VALID_URL = r'''(?x)
2296 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2299 annotation_id=annotation_[^&]+|
2305 attribution_link\?a=[^&]+
2311 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2312 'only_matching': True,
2314 'url': 'http://www.youtube.com/watch?',
2315 'only_matching': True,
2317 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2318 'only_matching': True,
2320 'url': 'https://www.youtube.com/watch?feature=foo',
2321 'only_matching': True,
2323 'url': 'https://www.youtube.com/watch?hl=en-GB',
2324 'only_matching': True,
2326 'url': 'https://www.youtube.com/watch?t=2372',
2327 'only_matching': True,
2330 def _real_extract(self, url):
2331 raise ExtractorError(
2332 'Did you forget to quote the URL? Remember that & is a meta '
2333 'character in most shells, so you want to put the URL in quotes, '
2335 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2336 ' or simply youtube-dl BaW_jenozKc .',
2340 class YoutubeTruncatedIDIE(InfoExtractor):
2341 IE_NAME = 'youtube:truncated_id'
2342 IE_DESC = False # Do not list
2343 _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2346 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2347 'only_matching': True,
2350 def _real_extract(self, url):
2351 video_id = self._match_id(url)
2352 raise ExtractorError(
2353 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),