3 from __future__
import unicode_literals
13 from .common
import InfoExtractor
, SearchInfoExtractor
14 from ..jsinterp
import JSInterpreter
15 from ..swfinterp
import SWFInterpreter
16 from ..compat
import (
20 compat_urllib_parse_unquote
,
21 compat_urllib_parse_unquote_plus
,
22 compat_urllib_parse_urlparse
,
31 get_element_by_attribute
,
48 class YoutubeBaseInfoExtractor(InfoExtractor
):
49 """Provide base functions for Youtube extractors"""
50 _LOGIN_URL
= 'https://accounts.google.com/ServiceLogin'
51 _TWOFACTOR_URL
= 'https://accounts.google.com/signin/challenge'
52 _NETRC_MACHINE
= 'youtube'
53 # If True it will raise an error if no login info is provided
54 _LOGIN_REQUIRED
= False
56 def _set_language(self
):
58 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
59 # YouTube sets the expire time to about two months
60 expire_time
=time
.time() + 2 * 30 * 24 * 3600)
62 def _ids_to_results(self
, ids
):
64 self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
69 Attempt to log in to YouTube.
70 True is returned if successful or skipped.
71 False is returned if login failed.
73 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
75 (username
, password
) = self
._get
_login
_info
()
76 # No authentication to be performed
78 if self
._LOGIN
_REQUIRED
:
79 raise ExtractorError('No login info available, needed for using %s.' % self
.IE_NAME
, expected
=True)
82 login_page
= self
._download
_webpage
(
83 self
._LOGIN
_URL
, None,
84 note
='Downloading login page',
85 errnote
='unable to fetch login page', fatal
=False)
86 if login_page
is False:
89 galx
= self
._search
_regex
(r
'(?s)<input.+?name="GALX".+?value="(.+?)"',
90 login_page
, 'Login GALX parameter')
94 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
99 'PersistentCookie': 'yes',
101 'bgresponse': 'js_disabled',
102 'checkConnection': '',
103 'checkedDomains': 'youtube',
110 'service': 'youtube',
115 login_data
= compat_urllib_parse
.urlencode(encode_dict(login_form_strs
)).encode('ascii')
117 req
= sanitized_Request(self
._LOGIN
_URL
, login_data
)
118 login_results
= self
._download
_webpage
(
120 note
='Logging in', errnote
='unable to log in', fatal
=False)
121 if login_results
is False:
124 if re
.search(r
'id="errormsg_0_Passwd"', login_results
) is not None:
125 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected
=True)
128 # TODO add SMS and phone call support - these require making a request and then prompting the user
130 if re
.search(r
'(?i)<form[^>]* id="challenge"', login_results
) is not None:
131 tfa_code
= self
._get
_tfa
_info
('2-step verification code')
134 self
._downloader
.report_warning(
135 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
136 '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
139 tfa_code
= remove_start(tfa_code
, 'G-')
141 tfa_form_strs
= self
._form
_hidden
_inputs
('challenge', login_results
)
143 tfa_form_strs
.update({
148 tfa_data
= compat_urllib_parse
.urlencode(encode_dict(tfa_form_strs
)).encode('ascii')
150 tfa_req
= sanitized_Request(self
._TWOFACTOR
_URL
, tfa_data
)
151 tfa_results
= self
._download
_webpage
(
153 note
='Submitting TFA code', errnote
='unable to submit tfa', fatal
=False)
155 if tfa_results
is False:
158 if re
.search(r
'(?i)<form[^>]* id="challenge"', tfa_results
) is not None:
159 self
._downloader
.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
161 if re
.search(r
'(?i)<form[^>]* id="gaia_loginform"', tfa_results
) is not None:
162 self
._downloader
.report_warning('unable to log in - did the page structure change?')
164 if re
.search(r
'smsauth-interstitial-reviewsettings', tfa_results
) is not None:
165 self
._downloader
.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
168 if re
.search(r
'(?i)<form[^>]* id="gaia_loginform"', login_results
) is not None:
169 self
._downloader
.report_warning('unable to log in: bad username or password')
173 def _real_initialize(self
):
174 if self
._downloader
is None:
177 if not self
._login
():
181 class YoutubeEntryListBaseInfoExtractor(InfoExtractor
):
182 # Extract entries from page with "Load more" button
183 def _entries(self
, page
, playlist_id
):
184 more_widget_html
= content_html
= page
185 for page_num
in itertools
.count(1):
186 for entry
in self
._process
_page
(content_html
):
189 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
193 more
= self
._download
_json
(
194 'https://youtube.com/%s' % mobj
.group('more'), playlist_id
,
195 'Downloading page #%s' % page_num
,
196 transform_source
=uppercase_escape
)
197 content_html
= more
['content_html']
198 if not content_html
.strip():
199 # Some webpages show a "Load more" button but they don't
202 more_widget_html
= more
['load_more_widget_html']
205 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor
):
206 def _process_page(self
, content
):
207 for video_id
, video_title
in self
.extract_videos_from_page(content
):
208 yield self
.url_result(video_id
, 'Youtube', video_id
, video_title
)
210 def extract_videos_from_page(self
, page
):
213 for mobj
in re
.finditer(self
._VIDEO
_RE
, page
):
214 # The link with index 0 is not the first video of the playlist (not sure if still actual)
215 if 'index' in mobj
.groupdict() and mobj
.group('id') == '0':
217 video_id
= mobj
.group('id')
218 video_title
= unescapeHTML(mobj
.group('title'))
220 video_title
= video_title
.strip()
222 idx
= ids_in_page
.index(video_id
)
223 if video_title
and not titles_in_page
[idx
]:
224 titles_in_page
[idx
] = video_title
226 ids_in_page
.append(video_id
)
227 titles_in_page
.append(video_title
)
228 return zip(ids_in_page
, titles_in_page
)
231 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor
):
232 def _process_page(self
, content
):
233 for playlist_id
in re
.findall(r
'href="/?playlist\?list=(.+?)"', content
):
234 yield self
.url_result(
235 'https://www.youtube.com/playlist?list=%s' % playlist_id
, 'YoutubePlaylist')
237 def _real_extract(self
, url
):
238 playlist_id
= self
._match
_id
(url
)
239 webpage
= self
._download
_webpage
(url
, playlist_id
)
240 title
= self
._og
_search
_title
(webpage
, fatal
=False)
241 return self
.playlist_result(self
._entries
(webpage
, playlist_id
), playlist_id
, title
)
244 class YoutubeIE(YoutubeBaseInfoExtractor
):
245 IE_DESC
= 'YouTube.com'
246 _VALID_URL
= r
"""(?x)^
248 (?:https?://|//) # http(s):// or protocol-independent URL
249 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
250 (?:www\.)?deturl\.com/www\.youtube\.com/|
251 (?:www\.)?pwnyoutube\.com/|
252 (?:www\.)?yourepeat\.com/|
253 tube\.majestyc\.net/|
254 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
255 (?:.*?\#/)? # handle anchor (#/) redirect urls
256 (?: # the various things that can precede the ID:
257 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
258 |(?: # or the v= param in all its forms
259 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
260 (?:\?|\#!?) # the params delimiter ? or # or #!
261 (?:.*?&)?? # any other preceding param (like /?s=tuff&v=xxxx)
266 youtu\.be| # just youtu.be/xxxx
267 vid\.plus # or vid.plus/xxxx
269 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
271 )? # all until now is optional -> you can pass the naked ID
272 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
273 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
274 (?(1).+)? # if we found the ID, everything can follow
276 _NEXT_URL_RE
= r
'[\?&]next_url=([^&]+)'
278 '5': {'ext': 'flv', 'width': 400, 'height': 240}
,
279 '6': {'ext': 'flv', 'width': 450, 'height': 270}
,
280 '13': {'ext': '3gp'}
,
281 '17': {'ext': '3gp', 'width': 176, 'height': 144}
,
282 '18': {'ext': 'mp4', 'width': 640, 'height': 360}
,
283 '22': {'ext': 'mp4', 'width': 1280, 'height': 720}
,
284 '34': {'ext': 'flv', 'width': 640, 'height': 360}
,
285 '35': {'ext': 'flv', 'width': 854, 'height': 480}
,
286 '36': {'ext': '3gp', 'width': 320, 'height': 240}
,
287 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080}
,
288 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072}
,
289 '43': {'ext': 'webm', 'width': 640, 'height': 360}
,
290 '44': {'ext': 'webm', 'width': 854, 'height': 480}
,
291 '45': {'ext': 'webm', 'width': 1280, 'height': 720}
,
292 '46': {'ext': 'webm', 'width': 1920, 'height': 1080}
,
293 '59': {'ext': 'mp4', 'width': 854, 'height': 480}
,
294 '78': {'ext': 'mp4', 'width': 854, 'height': 480}
,
298 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20}
,
299 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20}
,
300 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20}
,
301 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20}
,
302 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20}
,
303 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20}
,
304 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20}
,
306 # Apple HTTP Live Streaming
307 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}
,
308 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10}
,
309 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10}
,
310 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10}
,
311 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10}
,
312 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10}
,
313 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10}
,
316 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
317 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
318 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
319 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
320 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
321 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
322 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
323 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
324 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}
,
325 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'}
,
326 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'}
,
329 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'}
,
330 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'}
,
331 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'}
,
334 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
335 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
336 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
337 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
338 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
339 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'vp8', 'preference': -40}
,
340 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'vp9'}
,
341 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
342 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
343 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
344 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
345 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
346 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
347 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
348 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
349 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}
,
350 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}
,
351 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}
,
352 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}
,
353 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'vp9'}
,
354 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'vp9'}
,
357 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50}
,
358 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50}
,
360 # Dash webm audio with opus inside
361 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50}
,
362 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50}
,
363 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50}
,
366 '_rtmp': {'protocol': 'rtmp'}
,
372 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s&end=9',
376 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
377 'uploader': 'Philipp Hagemeister',
378 'uploader_id': 'phihag',
379 'upload_date': '20121002',
380 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
381 'categories': ['Science & Technology'],
382 'tags': ['youtube-dl'],
384 'dislike_count': int,
390 'url': 'http://www.youtube.com/watch?v=UxxajLWwzqY',
391 'note': 'Test generic use_cipher_signature video (#897)',
395 'upload_date': '20120506',
396 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
397 'description': 'md5:782e8651347686cba06e58f71ab51773',
398 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
399 'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
400 'iconic ep', 'iconic', 'love', 'it'],
401 'uploader': 'Icona Pop',
402 'uploader_id': 'IconaPop',
406 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
407 'note': 'Test VEVO video with age protection (#956)',
411 'upload_date': '20130703',
412 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
413 'description': 'md5:64249768eec3bc4276236606ea996373',
414 'uploader': 'justintimberlakeVEVO',
415 'uploader_id': 'justintimberlakeVEVO',
420 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
421 'note': 'Embed-only video (#1746)',
425 'upload_date': '20120608',
426 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
427 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
428 'uploader': 'SET India',
429 'uploader_id': 'setindia',
434 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&v=UxxajLWwzqY',
435 'note': 'Use the first video ID in the URL',
439 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
440 'uploader': 'Philipp Hagemeister',
441 'uploader_id': 'phihag',
442 'upload_date': '20121002',
443 '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 .',
444 'categories': ['Science & Technology'],
445 'tags': ['youtube-dl'],
447 'dislike_count': int,
450 'skip_download': True,
454 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
455 'note': '256k DASH audio (format 141) via DASH manifest',
459 'upload_date': '20121002',
460 'uploader_id': '8KVIDEO',
462 'uploader': '8KVIDEO',
463 'title': 'UHDTV TEST 8K VIDEO.mp4'
466 'youtube_include_dash_manifest': True,
470 # DASH manifest with encrypted signature
472 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
476 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
477 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
478 'uploader': 'AfrojackVEVO',
479 'uploader_id': 'AfrojackVEVO',
480 'upload_date': '20131011',
483 'youtube_include_dash_manifest': True,
487 # JS player signature function name containing $
489 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
493 'title': 'Taylor Swift - Shake It Off',
494 'description': 'md5:95f66187cd7c8b2c13eb78e1223b63c3',
495 'uploader': 'TaylorSwiftVEVO',
496 'uploader_id': 'TaylorSwiftVEVO',
497 'upload_date': '20140818',
500 'youtube_include_dash_manifest': True,
506 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
510 'upload_date': '20100909',
511 'uploader': 'The Amazing Atheist',
512 'uploader_id': 'TheAmazingAtheist',
513 'title': 'Burning Everyone\'s Koran',
514 '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',
517 # Normal age-gate video (No vevo, embed allowed)
519 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
523 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
524 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
525 'uploader': 'The Witcher',
526 'uploader_id': 'WitcherGame',
527 'upload_date': '20140605',
531 # Age-gate video with encrypted signature
533 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
537 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
538 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
539 'uploader': 'LloydVEVO',
540 'uploader_id': 'LloydVEVO',
541 'upload_date': '20110629',
545 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
547 'url': '__2ABJjxzNo',
551 'upload_date': '20100430',
552 'uploader_id': 'deadmau5',
553 'description': 'md5:12c56784b8032162bb936a5f76d55360',
554 'uploader': 'deadmau5',
555 'title': 'Deadmau5 - Some Chords (HD)',
557 'expected_warnings': [
558 'DASH manifest missing',
561 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
563 'url': 'lqQg6PlCWgI',
567 'upload_date': '20150827',
568 'uploader_id': 'olympic',
569 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
570 'uploader': 'Olympics',
571 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
574 'skip_download': 'requires avconv',
579 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
583 'stretched_ratio': 16 / 9.,
584 'upload_date': '20110310',
585 'uploader_id': 'AllenMeow',
586 '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',
588 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
591 # url_encoded_fmt_stream_map is empty string
593 'url': 'qEJwOuvDf7I',
597 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
599 'upload_date': '20150404',
600 'uploader_id': 'spbelect',
601 'uploader': 'Наблюдатели Петербурга',
604 'skip_download': 'requires avconv',
607 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
609 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
613 'title': 'md5:7b81415841e02ecd4313668cde88737a',
614 'description': 'md5:116377fd2963b81ec4ce64b542173306',
615 'upload_date': '20150625',
616 'uploader_id': 'dorappi2000',
617 'uploader': 'dorappi2000',
618 'formats': 'mincount:33',
621 # DASH manifest with segment_list
623 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
624 'md5': '8ce563a1d667b599d21064e982ab9e31',
628 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
629 'uploader': 'Airtek',
630 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
631 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
632 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
635 'youtube_include_dash_manifest': True,
636 'format': '135', # bestvideo
640 # Multifeed videos (multiple cameras), URL is for Main Camera
641 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
644 'title': 'teamPGP: Rocket League Noob Stream',
645 'description': 'md5:dc7872fb300e143831327f1bae3af010',
651 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
652 'description': 'md5:dc7872fb300e143831327f1bae3af010',
653 'upload_date': '20150721',
654 'uploader': 'Beer Games Beer',
655 'uploader_id': 'beergamesbeer',
661 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
662 'description': 'md5:dc7872fb300e143831327f1bae3af010',
663 'upload_date': '20150721',
664 'uploader': 'Beer Games Beer',
665 'uploader_id': 'beergamesbeer',
671 'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
672 'description': 'md5:dc7872fb300e143831327f1bae3af010',
673 'upload_date': '20150721',
674 'uploader': 'Beer Games Beer',
675 'uploader_id': 'beergamesbeer',
681 'title': 'teamPGP: Rocket League Noob Stream (zim)',
682 'description': 'md5:dc7872fb300e143831327f1bae3af010',
683 'upload_date': '20150721',
684 'uploader': 'Beer Games Beer',
685 'uploader_id': 'beergamesbeer',
689 'skip_download': True,
693 'url': 'http://vid.plus/FlRa-iH7PGw',
694 'only_matching': True,
697 # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
698 'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
702 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
703 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
704 'upload_date': '20151119',
705 'uploader_id': 'IronSoulElf',
706 'uploader': 'IronSoulElf',
709 'skip_download': True,
713 # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
714 'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
715 'only_matching': True,
718 # Video with yt:stretch=17:0
719 'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
723 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
724 'description': 'md5:ee18a25c350637c8faff806845bddee9',
725 'upload_date': '20151107',
726 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
727 'uploader': 'CH GAMER DROID',
730 'skip_download': True,
735 def __init__(self
, *args
, **kwargs
):
736 super(YoutubeIE
, self
).__init
__(*args
, **kwargs
)
737 self
._player
_cache
= {}
739 def report_video_info_webpage_download(self
, video_id
):
740 """Report attempt to download video info webpage."""
741 self
.to_screen('%s: Downloading video info webpage' % video_id
)
743 def report_information_extraction(self
, video_id
):
744 """Report attempt to extract video information."""
745 self
.to_screen('%s: Extracting video information' % video_id
)
747 def report_unavailable_format(self
, video_id
, format
):
748 """Report extracted video URL."""
749 self
.to_screen('%s: Format %s not available' % (video_id
, format
))
751 def report_rtmp_download(self
):
752 """Indicate the download will use the RTMP protocol."""
753 self
.to_screen('RTMP download detected')
755 def _signature_cache_id(self
, example_sig
):
756 """ Return a string representation of a signature """
757 return '.'.join(compat_str(len(part
)) for part
in example_sig
.split('.'))
759 def _extract_signature_function(self
, video_id
, player_url
, example_sig
):
761 r
'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|/base)?\.(?P<ext>[a-z]+)$',
764 raise ExtractorError('Cannot identify player %r' % player_url
)
765 player_type
= id_m
.group('ext')
766 player_id
= id_m
.group('id')
768 # Read from filesystem cache
769 func_id
= '%s_%s_%s' % (
770 player_type
, player_id
, self
._signature
_cache
_id
(example_sig
))
771 assert os
.path
.basename(func_id
) == func_id
773 cache_spec
= self
._downloader
.cache
.load('youtube-sigfuncs', func_id
)
774 if cache_spec
is not None:
775 return lambda s
: ''.join(s
[i
] for i
in cache_spec
)
778 'Downloading player %s' % player_url
779 if self
._downloader
.params
.get('verbose') else
780 'Downloading %s player %s' % (player_type
, player_id
)
782 if player_type
== 'js':
783 code
= self
._download
_webpage
(
784 player_url
, video_id
,
786 errnote
='Download of %s failed' % player_url
)
787 res
= self
._parse
_sig
_js
(code
)
788 elif player_type
== 'swf':
789 urlh
= self
._request
_webpage
(
790 player_url
, video_id
,
792 errnote
='Download of %s failed' % player_url
)
794 res
= self
._parse
_sig
_swf
(code
)
796 assert False, 'Invalid player type %r' % player_type
798 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
799 cache_res
= res(test_string
)
800 cache_spec
= [ord(c
) for c
in cache_res
]
802 self
._downloader
.cache
.store('youtube-sigfuncs', func_id
, cache_spec
)
805 def _print_sig_code(self
, func
, example_sig
):
806 def gen_sig_code(idxs
):
807 def _genslice(start
, end
, step
):
808 starts
= '' if start
== 0 else str(start
)
809 ends
= (':%d' % (end
+ step
)) if end
+ step
>= 0 else ':'
810 steps
= '' if step
== 1 else (':%d' % step
)
811 return 's[%s%s%s]' % (starts
, ends
, steps
)
814 # Quelch pyflakes warnings - start will be set when step is set
815 start
= '(Never used)'
816 for i
, prev
in zip(idxs
[1:], idxs
[:-1]):
820 yield _genslice(start
, prev
, step
)
823 if i
- prev
in [-1, 1]:
832 yield _genslice(start
, i
, step
)
834 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
835 cache_res
= func(test_string
)
836 cache_spec
= [ord(c
) for c
in cache_res
]
837 expr_code
= ' + '.join(gen_sig_code(cache_spec
))
838 signature_id_tuple
= '(%s)' % (
839 ', '.join(compat_str(len(p
)) for p
in example_sig
.split('.')))
840 code
= ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
841 ' return %s\n') % (signature_id_tuple
, expr_code
)
842 self
.to_screen('Extracted signature function:\n' + code
)
844 def _parse_sig_js(self
, jscode
):
845 funcname
= self
._search
_regex
(
846 r
'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode
,
847 'Initial JS player signature function name')
849 jsi
= JSInterpreter(jscode
)
850 initial_function
= jsi
.extract_function(funcname
)
851 return lambda s
: initial_function([s
])
853 def _parse_sig_swf(self
, file_contents
):
854 swfi
= SWFInterpreter(file_contents
)
855 TARGET_CLASSNAME
= 'SignatureDecipher'
856 searched_class
= swfi
.extract_class(TARGET_CLASSNAME
)
857 initial_function
= swfi
.extract_function(searched_class
, 'decipher')
858 return lambda s
: initial_function([s
])
860 def _decrypt_signature(self
, s
, video_id
, player_url
, age_gate
=False):
861 """Turn the encrypted s field into a working signature"""
863 if player_url
is None:
864 raise ExtractorError('Cannot decrypt signature without player_url')
866 if player_url
.startswith('//'):
867 player_url
= 'https:' + player_url
869 player_id
= (player_url
, self
._signature
_cache
_id
(s
))
870 if player_id
not in self
._player
_cache
:
871 func
= self
._extract
_signature
_function
(
872 video_id
, player_url
, s
874 self
._player
_cache
[player_id
] = func
875 func
= self
._player
_cache
[player_id
]
876 if self
._downloader
.params
.get('youtube_print_sig_code'):
877 self
._print
_sig
_code
(func
, s
)
879 except Exception as e
:
880 tb
= traceback
.format_exc()
881 raise ExtractorError(
882 'Signature extraction failed: ' + tb
, cause
=e
)
884 def _get_subtitles(self
, video_id
, webpage
):
886 subs_doc
= self
._download
_xml
(
887 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
,
888 video_id
, note
=False)
889 except ExtractorError
as err
:
890 self
._downloader
.report_warning('unable to download video subtitles: %s' % compat_str(err
))
894 for track
in subs_doc
.findall('track'):
895 lang
= track
.attrib
['lang_code']
896 if lang
in sub_lang_list
:
899 for ext
in ['sbv', 'vtt', 'srt']:
900 params
= compat_urllib_parse
.urlencode({
904 'name': track
.attrib
['name'].encode('utf-8'),
907 'url': 'https://www.youtube.com/api/timedtext?' + params
,
910 sub_lang_list
[lang
] = sub_formats
911 if not sub_lang_list
:
912 self
._downloader
.report_warning('video doesn\'t have subtitles')
916 def _get_ytplayer_config(self
, video_id
, webpage
):
918 # User data may contain arbitrary character sequences that may affect
919 # JSON extraction with regex, e.g. when '};' is contained the second
920 # regex won't capture the whole JSON. Yet working around by trying more
921 # concrete regex first keeping in mind proper quoted string handling
922 # to be implemented in future that will replace this workaround (see
923 # https://github.com/rg3/youtube-dl/issues/7468,
924 # https://github.com/rg3/youtube-dl/pull/7599)
925 r
';ytplayer\.config\s*=\s*({.+?});ytplayer',
926 r
';ytplayer\.config\s*=\s*({.+?});',
928 config
= self
._search
_regex
(
929 patterns
, webpage
, 'ytplayer.config', default
=None)
931 return self
._parse
_json
(
932 uppercase_escape(config
), video_id
, fatal
=False)
934 def _get_automatic_captions(self
, video_id
, webpage
):
935 """We need the webpage for getting the captions url, pass it as an
936 argument to speed up the process."""
937 self
.to_screen('%s: Looking for automatic captions' % video_id
)
938 player_config
= self
._get
_ytplayer
_config
(video_id
, webpage
)
939 err_msg
= 'Couldn\'t find automatic captions for %s' % video_id
940 if not player_config
:
941 self
._downloader
.report_warning(err_msg
)
944 args
= player_config
['args']
945 caption_url
= args
['ttsurl']
946 timestamp
= args
['timestamp']
947 # We get the available subtitles
948 list_params
= compat_urllib_parse
.urlencode({
953 list_url
= caption_url
+ '&' + list_params
954 caption_list
= self
._download
_xml
(list_url
, video_id
)
955 original_lang_node
= caption_list
.find('track')
956 if original_lang_node
is None:
957 self
._downloader
.report_warning('Video doesn\'t have automatic captions')
959 original_lang
= original_lang_node
.attrib
['lang_code']
960 caption_kind
= original_lang_node
.attrib
.get('kind', '')
963 for lang_node
in caption_list
.findall('target'):
964 sub_lang
= lang_node
.attrib
['lang_code']
966 for ext
in ['sbv', 'vtt', 'srt']:
967 params
= compat_urllib_parse
.urlencode({
968 'lang': original_lang
,
972 'kind': caption_kind
,
975 'url': caption_url
+ '&' + params
,
978 sub_lang_list
[sub_lang
] = sub_formats
980 # An extractor error can be raise by the download process if there are
981 # no automatic captions but there are subtitles
982 except (KeyError, ExtractorError
):
983 self
._downloader
.report_warning(err_msg
)
987 def extract_id(cls
, url
):
988 mobj
= re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
)
990 raise ExtractorError('Invalid URL: %s' % url
)
991 video_id
= mobj
.group(2)
994 def _extract_from_m3u8(self
, manifest_url
, video_id
):
997 def _get_urls(_manifest
):
998 lines
= _manifest
.split('\n')
999 urls
= filter(lambda l
: l
and not l
.startswith('#'),
1002 manifest
= self
._download
_webpage
(manifest_url
, video_id
, 'Downloading formats manifest')
1003 formats_urls
= _get_urls(manifest
)
1004 for format_url
in formats_urls
:
1005 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
1006 url_map
[itag
] = format_url
1009 def _extract_annotations(self
, video_id
):
1010 url
= 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
1011 return self
._download
_webpage
(url
, video_id
, note
='Searching for annotations.', errnote
='Unable to download video annotations.')
1013 def _parse_dash_manifest(
1014 self
, video_id
, dash_manifest_url
, player_url
, age_gate
, fatal
=True):
1015 def decrypt_sig(mobj
):
1017 dec_s
= self
._decrypt
_signature
(s
, video_id
, player_url
, age_gate
)
1018 return '/signature/%s' % dec_s
1019 dash_manifest_url
= re
.sub(r
'/s/([a-fA-F0-9\.]+)', decrypt_sig
, dash_manifest_url
)
1020 dash_doc
= self
._download
_xml
(
1021 dash_manifest_url
, video_id
,
1022 note
='Downloading DASH manifest',
1023 errnote
='Could not download DASH manifest',
1026 if dash_doc
is False:
1030 for a
in dash_doc
.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
1031 mime_type
= a
.attrib
.get('mimeType')
1032 for r
in a
.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
1033 url_el
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
1036 if mime_type
== 'text/vtt':
1037 # TODO implement WebVTT downloading
1039 elif mime_type
.startswith('audio/') or mime_type
.startswith('video/'):
1040 segment_list
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
1041 format_id
= r
.attrib
['id']
1042 video_url
= url_el
.text
1043 filesize
= int_or_none(url_el
.attrib
.get('{http://youtube.com/yt/2012/10/10}contentLength'))
1045 'format_id': format_id
,
1047 'width': int_or_none(r
.attrib
.get('width')),
1048 'height': int_or_none(r
.attrib
.get('height')),
1049 'tbr': int_or_none(r
.attrib
.get('bandwidth'), 1000),
1050 'asr': int_or_none(r
.attrib
.get('audioSamplingRate')),
1051 'filesize': filesize
,
1052 'fps': int_or_none(r
.attrib
.get('frameRate')),
1054 if segment_list
is not None:
1056 'initialization_url': segment_list
.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib
['sourceURL'],
1057 'segment_urls': [segment
.attrib
.get('media') for segment
in segment_list
.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
1058 'protocol': 'http_dash_segments',
1061 existing_format
= next(
1062 fo
for fo
in formats
1063 if fo
['format_id'] == format_id
)
1064 except StopIteration:
1065 full_info
= self
._formats
.get(format_id
, {}).copy()
1067 codecs
= r
.attrib
.get('codecs')
1069 if full_info
.get('acodec') == 'none' and 'vcodec' not in full_info
:
1070 full_info
['vcodec'] = codecs
1071 elif full_info
.get('vcodec') == 'none' and 'acodec' not in full_info
:
1072 full_info
['acodec'] = codecs
1073 formats
.append(full_info
)
1075 existing_format
.update(f
)
1077 self
.report_warning('Unknown MIME type %s in DASH manifest' % mime_type
)
1080 def _real_extract(self
, url
):
1081 url
, smuggled_data
= unsmuggle_url(url
, {})
1084 'http' if self
._downloader
.params
.get('prefer_insecure', False)
1089 parsed_url
= compat_urllib_parse_urlparse(url
)
1090 for component
in [parsed_url
.fragment
, parsed_url
.query
]:
1091 query
= compat_parse_qs(component
)
1092 if start_time
is None and 't' in query
:
1093 start_time
= parse_duration(query
['t'][0])
1094 if start_time
is None and 'start' in query
:
1095 start_time
= parse_duration(query
['start'][0])
1096 if end_time
is None and 'end' in query
:
1097 end_time
= parse_duration(query
['end'][0])
1099 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1100 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
1102 url
= proto
+ '://www.youtube.com/' + compat_urllib_parse_unquote(mobj
.group(1)).lstrip('/')
1103 video_id
= self
.extract_id(url
)
1106 url
= proto
+ '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1107 video_webpage
= self
._download
_webpage
(url
, video_id
)
1109 # Attempt to extract SWF player URL
1110 mobj
= re
.search(r
'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
1111 if mobj
is not None:
1112 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
1118 def add_dash_mpd(video_info
):
1119 dash_mpd
= video_info
.get('dashmpd')
1120 if dash_mpd
and dash_mpd
[0] not in dash_mpds
:
1121 dash_mpds
.append(dash_mpd
[0])
1124 embed_webpage
= None
1126 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
1128 # We simulate the access to the video from www.youtube.com/v/{video_id}
1129 # this can be viewed without login into Youtube
1130 url
= proto
+ '://www.youtube.com/embed/%s' % video_id
1131 embed_webpage
= self
._download
_webpage
(url
, video_id
, 'Downloading embed webpage')
1132 data
= compat_urllib_parse
.urlencode({
1133 'video_id': video_id
,
1134 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
1135 'sts': self
._search
_regex
(
1136 r
'"sts"\s*:\s*(\d+)', embed_webpage
, 'sts', default
=''),
1138 video_info_url
= proto
+ '://www.youtube.com/get_video_info?' + data
1139 video_info_webpage
= self
._download
_webpage
(
1140 video_info_url
, video_id
,
1141 note
='Refetching age-gated info webpage',
1142 errnote
='unable to download video info webpage')
1143 video_info
= compat_parse_qs(video_info_webpage
)
1144 add_dash_mpd(video_info
)
1148 # Try looking directly into the video webpage
1149 ytplayer_config
= self
._get
_ytplayer
_config
(video_id
, video_webpage
)
1151 args
= ytplayer_config
['args']
1152 if args
.get('url_encoded_fmt_stream_map'):
1153 # Convert to the same format returned by compat_parse_qs
1154 video_info
= dict((k
, [v
]) for k
, v
in args
.items())
1155 add_dash_mpd(video_info
)
1156 if args
.get('livestream') == '1' or args
.get('live_playback') == 1:
1158 if not video_info
or self
._downloader
.params
.get('youtube_include_dash_manifest', True):
1159 # We also try looking in get_video_info since it may contain different dashmpd
1160 # URL that points to a DASH manifest with possibly different itag set (some itags
1161 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1162 # manifest pointed by get_video_info's dashmpd).
1163 # The general idea is to take a union of itags of both DASH manifests (for example
1164 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
1165 self
.report_video_info_webpage_download(video_id
)
1166 for el_type
in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
1168 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
1169 % (proto
, video_id
, el_type
))
1170 video_info_webpage
= self
._download
_webpage
(
1172 video_id
, note
=False,
1173 errnote
='unable to download video info webpage')
1174 get_video_info
= compat_parse_qs(video_info_webpage
)
1175 if get_video_info
.get('use_cipher_signature') != ['True']:
1176 add_dash_mpd(get_video_info
)
1178 video_info
= get_video_info
1179 if 'token' in get_video_info
:
1180 # Different get_video_info requests may report different results, e.g.
1181 # some may report video unavailability, but some may serve it without
1182 # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
1183 # the original webpage as well as el=info and el=embedded get_video_info
1184 # requests report video unavailability due to geo restriction while
1185 # el=detailpage succeeds and returns valid data). This is probably
1186 # due to YouTube measures against IP ranges of hosting providers.
1187 # Working around by preferring the first succeeded video_info containing
1188 # the token if no such video_info yet was found.
1189 if 'token' not in video_info
:
1190 video_info
= get_video_info
1192 if 'token' not in video_info
:
1193 if 'reason' in video_info
:
1194 if 'The uploader has not made this video available in your country.' in video_info
['reason']:
1195 regions_allowed
= self
._html
_search
_meta
('regionsAllowed', video_webpage
, default
=None)
1197 raise ExtractorError('YouTube said: This video is available in %s only' % (
1198 ', '.join(map(ISO3166Utils
.short2full
, regions_allowed
.split(',')))),
1200 raise ExtractorError(
1201 'YouTube said: %s' % video_info
['reason'][0],
1202 expected
=True, video_id
=video_id
)
1204 raise ExtractorError(
1205 '"token" parameter not in video info for unknown reason',
1209 if 'title' in video_info
:
1210 video_title
= video_info
['title'][0]
1212 self
._downloader
.report_warning('Unable to extract video title')
1216 video_description
= get_element_by_id("eow-description", video_webpage
)
1217 if video_description
:
1218 video_description
= re
.sub(r
'''(?x)
1220 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1222 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1223 class="yt-uix-redirect-link"\s*>
1226 ''', r
'\1', video_description
)
1227 video_description
= clean_html(video_description
)
1229 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
1231 video_description
= unescapeHTML(fd_mobj
.group(1))
1233 video_description
= ''
1235 if 'multifeed_metadata_list' in video_info
and not smuggled_data
.get('force_singlefeed', False):
1236 if not self
._downloader
.params
.get('noplaylist'):
1239 multifeed_metadata_list
= compat_urllib_parse_unquote_plus(video_info
['multifeed_metadata_list'][0])
1240 for feed
in multifeed_metadata_list
.split(','):
1241 feed_data
= compat_parse_qs(feed
)
1243 '_type': 'url_transparent',
1244 'ie_key': 'Youtube',
1246 '%s://www.youtube.com/watch?v=%s' % (proto
, feed_data
['id'][0]),
1247 {'force_singlefeed': True}
),
1248 'title': '%s (%s)' % (video_title
, feed_data
['title'][0]),
1250 feed_ids
.append(feed_data
['id'][0])
1252 'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1253 % (', '.join(feed_ids
), video_id
))
1254 return self
.playlist_result(entries
, video_id
, video_title
, video_description
)
1255 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1257 if 'view_count' in video_info
:
1258 view_count
= int(video_info
['view_count'][0])
1262 # Check for "rental" videos
1263 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
1264 raise ExtractorError('"rental" videos not supported')
1266 # Start extracting information
1267 self
.report_information_extraction(video_id
)
1270 if 'author' not in video_info
:
1271 raise ExtractorError('Unable to extract uploader name')
1272 video_uploader
= compat_urllib_parse_unquote_plus(video_info
['author'][0])
1275 video_uploader_id
= None
1276 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
1277 if mobj
is not None:
1278 video_uploader_id
= mobj
.group(1)
1280 self
._downloader
.report_warning('unable to extract uploader nickname')
1283 # We try first to get a high quality image:
1284 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
1285 video_webpage
, re
.DOTALL
)
1286 if m_thumb
is not None:
1287 video_thumbnail
= m_thumb
.group(1)
1288 elif 'thumbnail_url' not in video_info
:
1289 self
._downloader
.report_warning('unable to extract video thumbnail')
1290 video_thumbnail
= None
1291 else: # don't panic if we can't find it
1292 video_thumbnail
= compat_urllib_parse_unquote_plus(video_info
['thumbnail_url'][0])
1295 upload_date
= self
._html
_search
_meta
(
1296 'datePublished', video_webpage
, 'upload date', default
=None)
1298 upload_date
= self
._search
_regex
(
1299 [r
'(?s)id="eow-date.*?>(.*?)</span>',
1300 r
'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1301 video_webpage
, 'upload date', default
=None)
1303 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
1304 upload_date
= unified_strdate(upload_date
)
1306 m_cat_container
= self
._search
_regex
(
1307 r
'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1308 video_webpage
, 'categories', default
=None)
1310 category
= self
._html
_search
_regex
(
1311 r
'(?s)<a[^<]+>(.*?)</a>', m_cat_container
, 'category',
1313 video_categories
= None if category
is None else [category
]
1315 video_categories
= None
1318 unescapeHTML(m
.group('content'))
1319 for m
in re
.finditer(self
._meta
_regex
('og:video:tag'), video_webpage
)]
1321 def _extract_count(count_name
):
1322 return str_to_int(self
._search
_regex
(
1323 r
'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1324 % re
.escape(count_name
),
1325 video_webpage
, count_name
, default
=None))
1327 like_count
= _extract_count('like')
1328 dislike_count
= _extract_count('dislike')
1331 video_subtitles
= self
.extract_subtitles(video_id
, video_webpage
)
1332 automatic_captions
= self
.extract_automatic_captions(video_id
, video_webpage
)
1334 if 'length_seconds' not in video_info
:
1335 self
._downloader
.report_warning('unable to extract video duration')
1336 video_duration
= None
1338 video_duration
= int(compat_urllib_parse_unquote_plus(video_info
['length_seconds'][0]))
1341 video_annotations
= None
1342 if self
._downloader
.params
.get('writeannotations', False):
1343 video_annotations
= self
._extract
_annotations
(video_id
)
1345 def _map_to_format_list(urlmap
):
1347 for itag
, video_real_url
in urlmap
.items():
1350 'url': video_real_url
,
1351 'player_url': player_url
,
1353 if itag
in self
._formats
:
1354 dct
.update(self
._formats
[itag
])
1358 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
1359 self
.report_rtmp_download()
1361 'format_id': '_rtmp',
1363 'url': video_info
['conn'][0],
1364 'player_url': player_url
,
1366 elif len(video_info
.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info
.get('adaptive_fmts', [''])[0]) >= 1:
1367 encoded_url_map
= video_info
.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info
.get('adaptive_fmts', [''])[0]
1368 if 'rtmpe%3Dyes' in encoded_url_map
:
1369 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
1371 for url_data_str
in encoded_url_map
.split(','):
1372 url_data
= compat_parse_qs(url_data_str
)
1373 if 'itag' not in url_data
or 'url' not in url_data
:
1375 format_id
= url_data
['itag'][0]
1376 url
= url_data
['url'][0]
1378 if 'sig' in url_data
:
1379 url
+= '&signature=' + url_data
['sig'][0]
1380 elif 's' in url_data
:
1381 encrypted_sig
= url_data
['s'][0]
1382 ASSETS_RE
= r
'"assets":.+?"js":\s*("[^"]+")'
1384 jsplayer_url_json
= self
._search
_regex
(
1386 embed_webpage
if age_gate
else video_webpage
,
1387 'JS player URL (1)', default
=None)
1388 if not jsplayer_url_json
and not age_gate
:
1389 # We need the embed website after all
1390 if embed_webpage
is None:
1391 embed_url
= proto
+ '://www.youtube.com/embed/%s' % video_id
1392 embed_webpage
= self
._download
_webpage
(
1393 embed_url
, video_id
, 'Downloading embed webpage')
1394 jsplayer_url_json
= self
._search
_regex
(
1395 ASSETS_RE
, embed_webpage
, 'JS player URL')
1397 player_url
= json
.loads(jsplayer_url_json
)
1398 if player_url
is None:
1399 player_url_json
= self
._search
_regex
(
1400 r
'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1401 video_webpage
, 'age gate player URL')
1402 player_url
= json
.loads(player_url_json
)
1404 if self
._downloader
.params
.get('verbose'):
1405 if player_url
is None:
1406 player_version
= 'unknown'
1407 player_desc
= 'unknown'
1409 if player_url
.endswith('swf'):
1410 player_version
= self
._search
_regex
(
1411 r
'-(.+?)(?:/watch_as3)?\.swf$', player_url
,
1412 'flash player', fatal
=False)
1413 player_desc
= 'flash player %s' % player_version
1415 player_version
= self
._search
_regex
(
1416 [r
'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js', r
'(?:www|player)-([^/]+)/base\.js'],
1418 'html5 player', fatal
=False)
1419 player_desc
= 'html5 player %s' % player_version
1421 parts_sizes
= self
._signature
_cache
_id
(encrypted_sig
)
1422 self
.to_screen('{%s} signature length %s, %s' %
1423 (format_id
, parts_sizes
, player_desc
))
1425 signature
= self
._decrypt
_signature
(
1426 encrypted_sig
, video_id
, player_url
, age_gate
)
1427 url
+= '&signature=' + signature
1428 if 'ratebypass' not in url
:
1429 url
+= '&ratebypass=yes'
1431 # Some itags are not included in DASH manifest thus corresponding formats will
1432 # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
1433 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
1434 mobj
= re
.search(r
'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data
.get('size', [''])[0])
1435 width
, height
= (int(mobj
.group('width')), int(mobj
.group('height'))) if mobj
else (None, None)
1437 'format_id': format_id
,
1439 'player_url': player_url
,
1440 'filesize': int_or_none(url_data
.get('clen', [None])[0]),
1441 'tbr': float_or_none(url_data
.get('bitrate', [None])[0], 1000),
1444 'fps': int_or_none(url_data
.get('fps', [None])[0]),
1445 'format_note': url_data
.get('quality_label', [None])[0] or url_data
.get('quality', [None])[0],
1447 type_
= url_data
.get('type', [None])[0]
1449 type_split
= type_
.split(';')
1450 kind_ext
= type_split
[0].split('/')
1451 if len(kind_ext
) == 2:
1452 kind
, ext
= kind_ext
1454 if kind
in ('audio', 'video'):
1456 for mobj
in re
.finditer(
1457 r
'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?
)(?P
<val
>.+?
)(?P
=quote
)(?
:;|$
)', type_):
1458 if mobj.group('key
') == 'codecs
':
1459 codecs = mobj.group('val
')
1462 codecs = codecs.split(',')
1463 if len(codecs) == 2:
1464 acodec, vcodec = codecs[0], codecs[1]
1466 acodec, vcodec = (codecs[0], 'none
') if kind == 'audio
' else ('none
', codecs[0])
1471 if format_id in self._formats:
1472 dct.update(self._formats[format_id])
1474 elif video_info.get('hlsvp
'):
1475 manifest_url = video_info['hlsvp
'][0]
1476 url_map = self._extract_from_m3u8(manifest_url, video_id)
1477 formats = _map_to_format_list(url_map)
1479 raise ExtractorError('no conn
, hlsvp
or url_encoded_fmt_stream_map information found
in video info
')
1481 # Look for the DASH manifest
1482 if self._downloader.params.get('youtube_include_dash_manifest
', True):
1483 dash_mpd_fatal = True
1484 for dash_manifest_url in dash_mpds:
1487 for df in self._parse_dash_manifest(
1488 video_id, dash_manifest_url, player_url, age_gate, dash_mpd_fatal):
1489 # Do not overwrite DASH format found in some previous DASH manifest
1490 if df['format_id
'] not in dash_formats:
1491 dash_formats[df['format_id
']] = df
1492 # Additional DASH manifests may end up in HTTP Error 403 therefore
1493 # allow them to fail without bug report message if we already have
1494 # some DASH manifest succeeded. This is temporary workaround to reduce
1495 # burst of bug reports until we figure out the reason and whether it
1496 # can be fixed at all.
1497 dash_mpd_fatal = False
1498 except (ExtractorError, KeyError) as e:
1499 self.report_warning(
1500 'Skipping DASH manifest
: %r' % e, video_id)
1502 # Remove the formats we found through non-DASH, they
1503 # contain less info and it can be wrong, because we use
1504 # fixed values (for example the resolution). See
1505 # https://github.com/rg3/youtube-dl/issues/5774 for an
1507 formats = [f for f in formats if f['format_id
'] not in dash_formats.keys()]
1508 formats.extend(dash_formats.values())
1510 # Check for malformed aspect ratio
1511 stretched_m = re.search(
1512 r'<meta\s
+property="og:video:tag".*?content
="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1515 w = float(stretched_m.group('w
'))
1516 h = float(stretched_m.group('h
'))
1520 if f.get('vcodec
') != 'none
':
1521 f['stretched_ratio
'] = ratio
1523 self._sort_formats(formats)
1527 'uploader
': video_uploader,
1528 'uploader_id
': video_uploader_id,
1529 'upload_date
': upload_date,
1530 'title
': video_title,
1531 'thumbnail
': video_thumbnail,
1532 'description
': video_description,
1533 'categories
': video_categories,
1535 'subtitles
': video_subtitles,
1536 'automatic_captions
': automatic_captions,
1537 'duration
': video_duration,
1538 'age_limit
': 18 if age_gate else 0,
1539 'annotations
': video_annotations,
1540 'webpage_url
': proto + '://www
.youtube
.com
/watch?v
=%s' % video_id,
1541 'view_count
': view_count,
1542 'like_count
': like_count,
1543 'dislike_count
': dislike_count,
1544 'average_rating
': float_or_none(video_info.get('avg_rating
', [None])[0]),
1547 'start_time
': start_time,
1548 'end_time
': end_time,
1552 class YoutubePlaylistIE(YoutubeBaseInfoExtractor, YoutubePlaylistBaseInfoExtractor):
1553 IE_DESC = 'YouTube
.com playlists
'
1554 _VALID_URL = r"""(?x)(?:
1559 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1560 \? (?:.*?&)*? (?:p|a|list)=
1564 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1565 # Top tracks, they can also include dots
1570 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1572 _TEMPLATE_URL = 'https
://www
.youtube
.com
/playlist?
list=%s'
1573 _VIDEO_RE = r'href
="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index
=(?P
<index
>\d
+)(?
:[^
>]+>(?P
<title
>[^
<]+))?
'
1574 IE_NAME = 'youtube
:playlist
'
1576 'url
': 'https
://www
.youtube
.com
/playlist?
list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re
',
1578 'title
': 'ytdl test PL
',
1579 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re
',
1581 'playlist_count
': 3,
1583 'url
': 'https
://www
.youtube
.com
/playlist?
list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx
',
1585 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx
',
1586 'title
': 'YDL_Empty_List
',
1588 'playlist_count
': 0,
1590 'note
': 'Playlist
with deleted
videos (#651). As a bonus, the video #51 is also twice in this list.',
1591 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1593 'title': '29C3: Not my department',
1594 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1596 'playlist_count': 95,
1598 'note': 'issue #673',
1599 'url': 'PLBB231211A4F62143',
1601 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1602 'id': 'PLBB231211A4F62143',
1604 'playlist_mincount': 26,
1606 'note': 'Large playlist',
1607 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1609 'title': 'Uploads from Cauchemar',
1610 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1612 'playlist_mincount': 799,
1614 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1616 'title': 'YDL_safe_search',
1617 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1619 'playlist_count': 2,
1622 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1623 'playlist_count': 4,
1626 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1629 'note': 'Embedded SWF player',
1630 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1631 'playlist_count': 4,
1634 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1637 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1638 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1640 'title': 'Uploads from Interstellar Movie',
1641 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1643 'playlist_mincout': 21,
1646 def _real_initialize(self
):
1649 def _extract_mix(self
, playlist_id
):
1650 # The mixes are generated from a single video
1651 # the id of the playlist is just 'RD' + video_id
1652 url
= 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id
[-11:], playlist_id
)
1653 webpage
= self
._download
_webpage
(
1654 url
, playlist_id
, 'Downloading Youtube mix')
1655 search_title
= lambda class_name
: get_element_by_attribute('class', class_name
, webpage
)
1657 search_title('playlist-title') or
1658 search_title('title long-title') or
1659 search_title('title'))
1660 title
= clean_html(title_span
)
1661 ids
= orderedSet(re
.findall(
1662 r
'''(?xs)data-video-username=".*?".*?
1663 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re
.escape(playlist_id
),
1665 url_results
= self
._ids
_to
_results
(ids
)
1667 return self
.playlist_result(url_results
, playlist_id
, title
)
1669 def _extract_playlist(self
, playlist_id
):
1670 url
= self
._TEMPLATE
_URL
% playlist_id
1671 page
= self
._download
_webpage
(url
, playlist_id
)
1673 for match
in re
.findall(r
'<div class="yt-alert-message">([^<]+)</div>', page
):
1674 match
= match
.strip()
1675 # Check if the playlist exists or is private
1676 if re
.match(r
'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match
):
1677 raise ExtractorError(
1678 'The playlist doesn\'t exist or is private, use --username or '
1679 '--netrc to access it.',
1681 elif re
.match(r
'[^<]*Invalid parameters[^<]*', match
):
1682 raise ExtractorError(
1683 'Invalid parameters. Maybe URL is incorrect.',
1685 elif re
.match(r
'[^<]*Choose your language[^<]*', match
):
1688 self
.report_warning('Youtube gives an alert message: ' + match
)
1690 playlist_title
= self
._html
_search
_regex
(
1691 r
'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
1694 return self
.playlist_result(self
._entries
(page
, playlist_id
), playlist_id
, playlist_title
)
1696 def _real_extract(self
, url
):
1697 # Extract playlist id
1698 mobj
= re
.match(self
._VALID
_URL
, url
)
1700 raise ExtractorError('Invalid URL: %s' % url
)
1701 playlist_id
= mobj
.group(1) or mobj
.group(2)
1703 # Check if it's a video-specific URL
1704 query_dict
= compat_urlparse
.parse_qs(compat_urlparse
.urlparse(url
).query
)
1705 if 'v' in query_dict
:
1706 video_id
= query_dict
['v'][0]
1707 if self
._downloader
.params
.get('noplaylist'):
1708 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1709 return self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1711 self
.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id
, video_id
))
1713 if playlist_id
.startswith('RD') or playlist_id
.startswith('UL'):
1714 # Mixes require a custom extraction process
1715 return self
._extract
_mix
(playlist_id
)
1717 return self
._extract
_playlist
(playlist_id
)
1720 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor
):
1721 IE_DESC
= 'YouTube.com channels'
1722 _VALID_URL
= r
'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1723 _TEMPLATE_URL
= 'https://www.youtube.com/channel/%s/videos'
1724 _VIDEO_RE
= r
'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
1725 IE_NAME
= 'youtube:channel'
1727 'note': 'paginated channel',
1728 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1729 'playlist_mincount': 91,
1731 'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
1732 'title': 'Uploads from lex will',
1735 'note': 'Age restricted channel',
1736 # from https://www.youtube.com/user/DeusExOfficial
1737 'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
1738 'playlist_mincount': 64,
1740 'id': 'UUs0ifCMCm1icqRbqhUINa0w',
1741 'title': 'Uploads from Deus Ex',
1745 def _real_extract(self
, url
):
1746 channel_id
= self
._match
_id
(url
)
1748 url
= self
._TEMPLATE
_URL
% channel_id
1750 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1751 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1752 # otherwise fallback on channel by page extraction
1753 channel_page
= self
._download
_webpage
(
1754 url
+ '?view=57', channel_id
,
1755 'Downloading channel page', fatal
=False)
1756 if channel_page
is False:
1757 channel_playlist_id
= False
1759 channel_playlist_id
= self
._html
_search
_meta
(
1760 'channelId', channel_page
, 'channel id', default
=None)
1761 if not channel_playlist_id
:
1762 channel_playlist_id
= self
._search
_regex
(
1763 r
'data-(?:channel-external-|yt)id="([^"]+)"',
1764 channel_page
, 'channel id', default
=None)
1765 if channel_playlist_id
and channel_playlist_id
.startswith('UC'):
1766 playlist_id
= 'UU' + channel_playlist_id
[2:]
1767 return self
.url_result(
1768 compat_urlparse
.urljoin(url
, '/playlist?list=%s' % playlist_id
), 'YoutubePlaylist')
1770 channel_page
= self
._download
_webpage
(url
, channel_id
, 'Downloading page #1')
1771 autogenerated
= re
.search(r
'''(?x)
1773 channel-header-autogenerated-label|
1774 yt-channel-title-autogenerated
1775 )[^"]*"''', channel_page
) is not None
1778 # The videos are contained in a single page
1779 # the ajax pages can't be used, they are empty
1782 video_id
, 'Youtube', video_id
=video_id
,
1783 video_title
=video_title
)
1784 for video_id
, video_title
in self
.extract_videos_from_page(channel_page
)]
1785 return self
.playlist_result(entries
, channel_id
)
1787 return self
.playlist_result(self
._entries
(channel_page
, channel_id
), channel_id
)
1790 class YoutubeUserIE(YoutubeChannelIE
):
1791 IE_DESC
= 'YouTube.com user videos (URL or "ytuser" keyword)'
1792 _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_-]+)'
1793 _TEMPLATE_URL
= 'https://www.youtube.com/user/%s/videos'
1794 IE_NAME
= 'youtube:user'
1797 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1798 'playlist_mincount': 320,
1800 'title': 'TheLinuxFoundation',
1803 'url': 'ytuser:phihag',
1804 'only_matching': True,
1808 def suitable(cls
, url
):
1809 # Don't return True if the url can be extracted with other youtube
1810 # extractor, the regex would is too permissive and it would match.
1811 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1812 if any(ie
.suitable(url
) for ie
in other_ies
):
1815 return super(YoutubeUserIE
, cls
).suitable(url
)
1818 class YoutubeUserPlaylistsIE(YoutubePlaylistsBaseInfoExtractor
):
1819 IE_DESC
= 'YouTube.com user playlists'
1820 _VALID_URL
= r
'https?://(?:\w+\.)?youtube\.com/user/(?P<id>[^/]+)/playlists'
1821 IE_NAME
= 'youtube:user:playlists'
1824 'url': 'http://www.youtube.com/user/ThirstForScience/playlists',
1825 'playlist_mincount': 4,
1827 'id': 'ThirstForScience',
1828 'title': 'Thirst for Science',
1831 # with "Load more" button
1832 'url': 'http://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
1833 'playlist_mincount': 70,
1836 'title': 'Игорь Клейнер',
1841 class YoutubeSearchIE(SearchInfoExtractor
, YoutubePlaylistIE
):
1842 IE_DESC
= 'YouTube.com searches'
1843 # there doesn't appear to be a real limit, for example if you search for
1844 # 'python' you get more than 8.000.000 results
1845 _MAX_RESULTS
= float('inf')
1846 IE_NAME
= 'youtube:search'
1847 _SEARCH_KEY
= 'ytsearch'
1848 _EXTRA_QUERY_ARGS
= {}
1851 def _get_n_results(self
, query
, n
):
1852 """Get a specified number of results for a query"""
1857 for pagenum
in itertools
.count(1):
1859 'search_query': query
.encode('utf-8'),
1863 url_query
.update(self
._EXTRA
_QUERY
_ARGS
)
1864 result_url
= 'https://www.youtube.com/results?' + compat_urllib_parse
.urlencode(url_query
)
1865 data
= self
._download
_json
(
1866 result_url
, video_id
='query "%s"' % query
,
1867 note
='Downloading page %s' % pagenum
,
1868 errnote
='Unable to download API page')
1869 html_content
= data
[1]['body']['content']
1871 if 'class="search-message' in html_content
:
1872 raise ExtractorError(
1873 '[youtube] No video results', expected
=True)
1875 new_videos
= self
._ids
_to
_results
(orderedSet(re
.findall(
1876 r
'href="/watch\?v=(.{11})', html_content
)))
1877 videos
+= new_videos
1878 if not new_videos
or len(videos
) > limit
:
1883 return self
.playlist_result(videos
, query
)
1886 class YoutubeSearchDateIE(YoutubeSearchIE
):
1887 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1888 _SEARCH_KEY
= 'ytsearchdate'
1889 IE_DESC
= 'YouTube.com searches, newest videos first'
1890 _EXTRA_QUERY_ARGS
= {'search_sort': 'video_date_uploaded'}
1893 class YoutubeSearchURLIE(InfoExtractor
):
1894 IE_DESC
= 'YouTube.com search URLs'
1895 IE_NAME
= 'youtube:search_url'
1896 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1898 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1899 'playlist_mincount': 5,
1901 'title': 'youtube-dl test video',
1905 def _real_extract(self
, url
):
1906 mobj
= re
.match(self
._VALID
_URL
, url
)
1907 query
= compat_urllib_parse_unquote_plus(mobj
.group('query'))
1909 webpage
= self
._download
_webpage
(url
, query
)
1910 result_code
= self
._search
_regex
(
1911 r
'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage
, 'result HTML')
1913 part_codes
= re
.findall(
1914 r
'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code
)
1916 for part_code
in part_codes
:
1917 part_title
= self
._html
_search
_regex
(
1918 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1919 part_url_snippet
= self
._html
_search
_regex
(
1920 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1921 part_url
= compat_urlparse
.urljoin(
1922 'https://www.youtube.com/', part_url_snippet
)
1926 'title': part_title
,
1930 '_type': 'playlist',
1936 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor
):
1937 IE_DESC
= 'YouTube.com (multi-season) shows'
1938 _VALID_URL
= r
'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1939 IE_NAME
= 'youtube:show'
1941 'url': 'https://www.youtube.com/show/airdisasters',
1942 'playlist_mincount': 5,
1944 'id': 'airdisasters',
1945 'title': 'Air Disasters',
1949 def _real_extract(self
, url
):
1950 playlist_id
= self
._match
_id
(url
)
1951 return super(YoutubeShowIE
, self
)._real
_extract
(
1952 'https://www.youtube.com/show/%s/playlists' % playlist_id
)
1955 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1957 Base class for feed extractors
1958 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1960 _LOGIN_REQUIRED
= True
1964 return 'youtube:%s' % self
._FEED
_NAME
1966 def _real_initialize(self
):
1969 def _real_extract(self
, url
):
1970 page
= self
._download
_webpage
(
1971 'https://www.youtube.com/feed/%s' % self
._FEED
_NAME
, self
._PLAYLIST
_TITLE
)
1973 # The extraction process is the same as for playlists, but the regex
1974 # for the video ids doesn't contain an index
1976 more_widget_html
= content_html
= page
1977 for page_num
in itertools
.count(1):
1978 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
1980 # 'recommended' feed has infinite 'load more' and each new portion spins
1981 # the same videos in (sometimes) slightly different order, so we'll check
1982 # for unicity and break when portion has no new videos
1983 new_ids
= filter(lambda video_id
: video_id
not in ids
, orderedSet(matches
))
1989 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1993 more
= self
._download
_json
(
1994 'https://youtube.com/%s' % mobj
.group('more'), self
._PLAYLIST
_TITLE
,
1995 'Downloading page #%s' % page_num
,
1996 transform_source
=uppercase_escape
)
1997 content_html
= more
['content_html']
1998 more_widget_html
= more
['load_more_widget_html']
2000 return self
.playlist_result(
2001 self
._ids
_to
_results
(ids
), playlist_title
=self
._PLAYLIST
_TITLE
)
2004 class YoutubeWatchLaterIE(YoutubePlaylistIE
):
2005 IE_NAME
= 'youtube:watchlater'
2006 IE_DESC
= 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
2007 _VALID_URL
= r
'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
2009 _TESTS
= [] # override PlaylistIE tests
2011 def _real_extract(self
, url
):
2012 return self
._extract
_playlist
('WL')
2015 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
2016 IE_NAME
= 'youtube:favorites'
2017 IE_DESC
= 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
2018 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
2019 _LOGIN_REQUIRED
= True
2021 def _real_extract(self
, url
):
2022 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
2023 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, 'favourites playlist id')
2024 return self
.url_result(playlist_id
, 'YoutubePlaylist')
2027 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
2028 IE_DESC
= 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
2029 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
2030 _FEED_NAME
= 'recommended'
2031 _PLAYLIST_TITLE
= 'Youtube Recommended videos'
2034 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor
):
2035 IE_DESC
= 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
2036 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
2037 _FEED_NAME
= 'subscriptions'
2038 _PLAYLIST_TITLE
= 'Youtube Subscriptions'
2041 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
2042 IE_DESC
= 'Youtube watch history, ":ythistory" for short (requires authentication)'
2043 _VALID_URL
= 'https?://www\.youtube\.com/feed/history|:ythistory'
2044 _FEED_NAME
= 'history'
2045 _PLAYLIST_TITLE
= 'Youtube History'
2048 class YoutubeTruncatedURLIE(InfoExtractor
):
2049 IE_NAME
= 'youtube:truncated_url'
2050 IE_DESC
= False # Do not list
2051 _VALID_URL
= r
'''(?x)
2053 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
2056 annotation_id=annotation_[^&]+|
2062 attribution_link\?a=[^&]+
2068 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
2069 'only_matching': True,
2071 'url': 'http://www.youtube.com/watch?',
2072 'only_matching': True,
2074 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
2075 'only_matching': True,
2077 'url': 'https://www.youtube.com/watch?feature=foo',
2078 'only_matching': True,
2080 'url': 'https://www.youtube.com/watch?hl=en-GB',
2081 'only_matching': True,
2083 'url': 'https://www.youtube.com/watch?t=2372',
2084 'only_matching': True,
2087 def _real_extract(self
, url
):
2088 raise ExtractorError(
2089 'Did you forget to quote the URL? Remember that & is a meta '
2090 'character in most shells, so you want to put the URL in quotes, '
2092 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
2093 ' or simply youtube-dl BaW_jenozKc .',
2097 class YoutubeTruncatedIDIE(InfoExtractor
):
2098 IE_NAME
= 'youtube:truncated_id'
2099 IE_DESC
= False # Do not list
2100 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
2103 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
2104 'only_matching': True,
2107 def _real_extract(self
, url
):
2108 video_id
= self
._match
_id
(url
)
2109 raise ExtractorError(
2110 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id
, url
),