3 from __future__
import unicode_literals
13 from .common
import InfoExtractor
, SearchInfoExtractor
14 from ..jsinterp
import JSInterpreter
15 from ..swfinterp
import SWFInterpreter
16 from ..compat
import (
20 compat_urllib_parse_unquote
,
21 compat_urllib_parse_unquote_plus
,
22 compat_urllib_parse_urlparse
,
23 compat_urllib_request
,
31 get_element_by_attribute
,
44 class YoutubeBaseInfoExtractor(InfoExtractor
):
45 """Provide base functions for Youtube extractors"""
46 _LOGIN_URL
= 'https://accounts.google.com/ServiceLogin'
47 _TWOFACTOR_URL
= 'https://accounts.google.com/SecondFactor'
48 _NETRC_MACHINE
= 'youtube'
49 # If True it will raise an error if no login info is provided
50 _LOGIN_REQUIRED
= False
52 def _set_language(self
):
54 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
55 # YouTube sets the expire time to about two months
56 expire_time
=time
.time() + 2 * 30 * 24 * 3600)
58 def _ids_to_results(self
, ids
):
60 self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
65 Attempt to log in to YouTube.
66 True is returned if successful or skipped.
67 False is returned if login failed.
69 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
71 (username
, password
) = self
._get
_login
_info
()
72 # No authentication to be performed
74 if self
._LOGIN
_REQUIRED
:
75 raise ExtractorError('No login info available, needed for using %s.' % self
.IE_NAME
, expected
=True)
78 login_page
= self
._download
_webpage
(
79 self
._LOGIN
_URL
, None,
80 note
='Downloading login page',
81 errnote
='unable to fetch login page', fatal
=False)
82 if login_page
is False:
85 galx
= self
._search
_regex
(r
'(?s)<input.+?name="GALX".+?value="(.+?)"',
86 login_page
, 'Login GALX parameter')
90 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
95 'PersistentCookie': 'yes',
97 'bgresponse': 'js_disabled',
98 'checkConnection': '',
99 'checkedDomains': 'youtube',
106 'service': 'youtube',
111 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
113 login_form
= dict((k
.encode('utf-8'), v
.encode('utf-8')) for k
, v
in login_form_strs
.items())
114 login_data
= compat_urllib_parse
.urlencode(login_form
).encode('ascii')
116 req
= compat_urllib_request
.Request(self
._LOGIN
_URL
, login_data
)
117 login_results
= self
._download
_webpage
(
119 note
='Logging in', errnote
='unable to log in', fatal
=False)
120 if login_results
is False:
123 if re
.search(r
'id="errormsg_0_Passwd"', login_results
) is not None:
124 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected
=True)
127 # TODO add SMS and phone call support - these require making a request and then prompting the user
129 if re
.search(r
'(?i)<form[^>]* id="gaia_secondfactorform"', login_results
) is not None:
130 tfa_code
= self
._get
_tfa
_info
()
133 self
._downloader
.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
134 self
._downloader
.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
137 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
139 match
= re
.search(r
'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
141 self._downloader.report_warning('Failed to get secTok
- did the page structure change?
')
142 secTok = match.group(1)
143 match = re.search(r'id="timeStmp"\n\s
+value
=\'(.+)\'/>', login_results, re.M | re.U)
145 self._downloader.report_warning('Failed to get timeStmp
- did the page structure change?
')
146 timeStmp = match.group(1)
149 'continue': 'https
://www
.youtube
.com
/signin?action_handle_signin
=true
&feature
=sign_in_button
&hl
=en_US
&nomobiletemp
=1',
151 'smsUserPin
': tfa_code,
152 'smsVerifyPin
': 'Verify
',
154 'PersistentCookie
': 'yes
',
155 'checkConnection
': '',
156 'checkedDomains
': 'youtube
',
159 'timeStmp
': timeStmp,
160 'service
': 'youtube
',
163 tfa_form = dict((k.encode('utf
-8'), v.encode('utf
-8')) for k, v in tfa_form_strs.items())
164 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii
')
166 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
167 tfa_results = self._download_webpage(
169 note='Submitting TFA code
', errnote='unable to submit tfa
', fatal=False)
171 if tfa_results is False:
174 if re.search(r'(?i
)<form
[^
>]* id="gaia_secondfactorform"', tfa_results) is not None:
175 self._downloader.report_warning('Two
-factor code expired
. Please
try again
, or use a one
-use backup code instead
.')
177 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', tfa_results) is not None:
178 self._downloader.report_warning('unable to log
in - did the page structure change?
')
180 if re.search(r'smsauth
-interstitial
-reviewsettings
', tfa_results) is not None:
181 self._downloader.report_warning('Your Google account has a security notice
. Please log
in on your web browser
, resolve the notice
, and try again
.')
184 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', login_results) is not None:
185 self._downloader.report_warning('unable to log
in: bad username
or password
')
189 def _real_initialize(self):
190 if self._downloader is None:
193 if not self._login():
197 class YoutubeIE(YoutubeBaseInfoExtractor):
198 IE_DESC = 'YouTube
.com
'
199 _VALID_URL = r"""(?x)^
201 (?:https?://|//) # http(s):// or protocol-independent URL
202 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
203 (?:www\.)?deturl\.com/www\.youtube\.com/|
204 (?:www\.)?pwnyoutube\.com/|
205 (?:www\.)?yourepeat\.com/|
206 tube\.majestyc\.net/|
207 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
208 (?:.*?\#/)? # handle anchor (#/) redirect urls
209 (?: # the various things that can precede the ID:
210 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
211 |(?: # or the v= param in all its forms
212 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
213 (?:\?|\#!?) # the params delimiter ? or # or #!
214 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
218 |youtu\.be/ # just youtu.be/xxxx
219 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
221 )? # all until now is optional -> you can pass the naked ID
222 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
223 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
224 (?(1).+)? # if we found the ID, everything can follow
226 _NEXT_URL_RE = r'[\?&]next_url
=([^
&]+)'
228 '5': {'ext': 'flv', 'width': 400, 'height': 240},
229 '6': {'ext': 'flv', 'width': 450, 'height': 270},
230 '13': {'ext': '3gp'},
231 '17': {'ext': '3gp', 'width': 176, 'height': 144},
232 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
233 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
234 '34': {'ext': 'flv', 'width': 640, 'height': 360},
235 '35': {'ext': 'flv', 'width': 854, 'height': 480},
236 '36': {'ext': '3gp', 'width': 320, 'height': 240},
237 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
238 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
239 '43': {'ext': 'webm', 'width': 640, 'height': 360},
240 '44': {'ext': 'webm', 'width': 854, 'height': 480},
241 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
242 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
243 '59': {'ext': 'mp4', 'width': 854, 'height': 480},
244 '78': {'ext': 'mp4', 'width': 854, 'height': 480},
248 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
249 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
250 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
251 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
252 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
253 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
254 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
256 # Apple HTTP Live Streaming
257 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
258 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
259 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
260 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
261 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
262 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
263 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
266 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
267 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
268 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
269 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
270 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
271 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
272 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
273 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
274 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
275 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
276 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
279 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
280 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
281 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
284 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
285 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
286 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
287 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
288 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
289 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
290 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
291 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
293 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
298 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
299 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
300 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
301 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
302 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
303 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
304 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
307 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
308 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
310 # Dash webm audio with opus inside
311 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
312 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
313 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
316 '_rtmp
': {'protocol': 'rtmp'},
322 'url
': 'http
://www
.youtube
.com
/watch?v
=BaW_jenozKcj
&t
=1s
&end
=9',
326 'title
': 'youtube
-dl test video
"\'/\\ä↭𝕐',
327 'uploader': 'Philipp Hagemeister',
328 'uploader_id': 'phihag',
329 'upload_date': '20121002',
330 '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 .',
331 'categories
': ['Science
& Technology
'],
333 'dislike_count
': int,
339 'url
': 'http
://www
.youtube
.com
/watch?v
=UxxajLWwzqY
',
340 'note
': 'Test generic use_cipher_signature
video (#897)',
344 'upload_date': '20120506',
345 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
346 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
347 'uploader': 'Icona Pop',
348 'uploader_id': 'IconaPop',
352 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
353 'note': 'Test VEVO video with age protection (#956)',
357 'upload_date': '20130703',
358 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
359 'description': 'md5:64249768eec3bc4276236606ea996373',
360 'uploader': 'justintimberlakeVEVO',
361 'uploader_id': 'justintimberlakeVEVO',
365 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
366 'note': 'Embed-only video (#1746)',
370 'upload_date': '20120608',
371 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
372 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
373 'uploader': 'SET India',
374 'uploader_id': 'setindia'
378 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
379 'note': '256k DASH audio (format 141) via DASH manifest',
383 'upload_date': '20121002',
384 'uploader_id': '8KVIDEO',
386 'uploader': '8KVIDEO',
387 'title': 'UHDTV TEST 8K VIDEO.mp4'
390 'youtube_include_dash_manifest': True,
394 # DASH manifest with encrypted signature
396 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
400 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
401 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
402 'uploader': 'AfrojackVEVO',
403 'uploader_id': 'AfrojackVEVO',
404 'upload_date': '20131011',
407 'youtube_include_dash_manifest': True,
411 # JS player signature function name containing $
413 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
417 'title': 'Taylor Swift - Shake It Off',
418 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
419 'uploader': 'TaylorSwiftVEVO',
420 'uploader_id': 'TaylorSwiftVEVO',
421 'upload_date': '20140818',
424 'youtube_include_dash_manifest': True,
430 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
434 'upload_date': '20100909',
435 'uploader': 'The Amazing Atheist',
436 'uploader_id': 'TheAmazingAtheist',
437 'title': 'Burning Everyone\'s Koran',
438 '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',
441 # Normal age-gate video (No vevo, embed allowed)
443 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
447 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
448 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
449 'uploader': 'The Witcher',
450 'uploader_id': 'WitcherGame',
451 'upload_date': '20140605',
454 # Age-gate video with encrypted signature
456 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
460 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
461 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
462 'uploader': 'LloydVEVO',
463 'uploader_id': 'LloydVEVO',
464 'upload_date': '20110629',
467 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
469 'url': '__2ABJjxzNo',
473 'upload_date': '20100430',
474 'uploader_id': 'deadmau5',
475 'description': 'md5:12c56784b8032162bb936a5f76d55360',
476 'uploader': 'deadmau5',
477 'title': 'Deadmau5 - Some Chords (HD)',
479 'expected_warnings': [
480 'DASH manifest missing',
483 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
485 'url': 'lqQg6PlCWgI',
489 'upload_date': '20120731',
490 'uploader_id': 'olympic',
491 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
492 'uploader': 'Olympics',
493 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
496 'skip_download': 'requires avconv',
501 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
505 'stretched_ratio': 16 / 9.,
506 'upload_date': '20110310',
507 'uploader_id': 'AllenMeow',
508 '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',
510 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
513 # url_encoded_fmt_stream_map is empty string
515 'url': 'qEJwOuvDf7I',
519 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
521 'upload_date': '20150404',
522 'uploader_id': 'spbelect',
523 'uploader': 'Наблюдатели Петербурга',
526 'skip_download': 'requires avconv',
529 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
531 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
535 'title': 'md5:7b81415841e02ecd4313668cde88737a',
536 'description': 'md5:116377fd2963b81ec4ce64b542173306',
537 'upload_date': '20150625',
538 'uploader_id': 'dorappi2000',
539 'uploader': 'dorappi2000',
540 'formats': 'mincount:33',
543 # DASH manifest with segment_list
545 'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
546 'md5': '8ce563a1d667b599d21064e982ab9e31',
550 'upload_date': '20150501', # According to '<meta itemprop="datePublished"', but in other places it's 20150510
551 'uploader': 'Airtek',
552 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
553 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
554 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
557 'youtube_include_dash_manifest': True,
558 'format': '135', # bestvideo
563 def __init__(self
, *args
, **kwargs
):
564 super(YoutubeIE
, self
).__init
__(*args
, **kwargs
)
565 self
._player
_cache
= {}
567 def report_video_info_webpage_download(self
, video_id
):
568 """Report attempt to download video info webpage."""
569 self
.to_screen('%s: Downloading video info webpage' % video_id
)
571 def report_information_extraction(self
, video_id
):
572 """Report attempt to extract video information."""
573 self
.to_screen('%s: Extracting video information' % video_id
)
575 def report_unavailable_format(self
, video_id
, format
):
576 """Report extracted video URL."""
577 self
.to_screen('%s: Format %s not available' % (video_id
, format
))
579 def report_rtmp_download(self
):
580 """Indicate the download will use the RTMP protocol."""
581 self
.to_screen('RTMP download detected')
583 def _signature_cache_id(self
, example_sig
):
584 """ Return a string representation of a signature """
585 return '.'.join(compat_str(len(part
)) for part
in example_sig
.split('.'))
587 def _extract_signature_function(self
, video_id
, player_url
, example_sig
):
589 r
'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
592 raise ExtractorError('Cannot identify player %r' % player_url
)
593 player_type
= id_m
.group('ext')
594 player_id
= id_m
.group('id')
596 # Read from filesystem cache
597 func_id
= '%s_%s_%s' % (
598 player_type
, player_id
, self
._signature
_cache
_id
(example_sig
))
599 assert os
.path
.basename(func_id
) == func_id
601 cache_spec
= self
._downloader
.cache
.load('youtube-sigfuncs', func_id
)
602 if cache_spec
is not None:
603 return lambda s
: ''.join(s
[i
] for i
in cache_spec
)
606 'Downloading player %s' % player_url
607 if self
._downloader
.params
.get('verbose') else
608 'Downloading %s player %s' % (player_type
, player_id
)
610 if player_type
== 'js':
611 code
= self
._download
_webpage
(
612 player_url
, video_id
,
614 errnote
='Download of %s failed' % player_url
)
615 res
= self
._parse
_sig
_js
(code
)
616 elif player_type
== 'swf':
617 urlh
= self
._request
_webpage
(
618 player_url
, video_id
,
620 errnote
='Download of %s failed' % player_url
)
622 res
= self
._parse
_sig
_swf
(code
)
624 assert False, 'Invalid player type %r' % player_type
626 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
627 cache_res
= res(test_string
)
628 cache_spec
= [ord(c
) for c
in cache_res
]
630 self
._downloader
.cache
.store('youtube-sigfuncs', func_id
, cache_spec
)
633 def _print_sig_code(self
, func
, example_sig
):
634 def gen_sig_code(idxs
):
635 def _genslice(start
, end
, step
):
636 starts
= '' if start
== 0 else str(start
)
637 ends
= (':%d' % (end
+ step
)) if end
+ step
>= 0 else ':'
638 steps
= '' if step
== 1 else (':%d' % step
)
639 return 's[%s%s%s]' % (starts
, ends
, steps
)
642 # Quelch pyflakes warnings - start will be set when step is set
643 start
= '(Never used)'
644 for i
, prev
in zip(idxs
[1:], idxs
[:-1]):
648 yield _genslice(start
, prev
, step
)
651 if i
- prev
in [-1, 1]:
660 yield _genslice(start
, i
, step
)
662 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
663 cache_res
= func(test_string
)
664 cache_spec
= [ord(c
) for c
in cache_res
]
665 expr_code
= ' + '.join(gen_sig_code(cache_spec
))
666 signature_id_tuple
= '(%s)' % (
667 ', '.join(compat_str(len(p
)) for p
in example_sig
.split('.')))
668 code
= ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
669 ' return %s\n') % (signature_id_tuple
, expr_code
)
670 self
.to_screen('Extracted signature function:\n' + code
)
672 def _parse_sig_js(self
, jscode
):
673 funcname
= self
._search
_regex
(
674 r
'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode
,
675 'Initial JS player signature function name')
677 jsi
= JSInterpreter(jscode
)
678 initial_function
= jsi
.extract_function(funcname
)
679 return lambda s
: initial_function([s
])
681 def _parse_sig_swf(self
, file_contents
):
682 swfi
= SWFInterpreter(file_contents
)
683 TARGET_CLASSNAME
= 'SignatureDecipher'
684 searched_class
= swfi
.extract_class(TARGET_CLASSNAME
)
685 initial_function
= swfi
.extract_function(searched_class
, 'decipher')
686 return lambda s
: initial_function([s
])
688 def _decrypt_signature(self
, s
, video_id
, player_url
, age_gate
=False):
689 """Turn the encrypted s field into a working signature"""
691 if player_url
is None:
692 raise ExtractorError('Cannot decrypt signature without player_url')
694 if player_url
.startswith('//'):
695 player_url
= 'https:' + player_url
697 player_id
= (player_url
, self
._signature
_cache
_id
(s
))
698 if player_id
not in self
._player
_cache
:
699 func
= self
._extract
_signature
_function
(
700 video_id
, player_url
, s
702 self
._player
_cache
[player_id
] = func
703 func
= self
._player
_cache
[player_id
]
704 if self
._downloader
.params
.get('youtube_print_sig_code'):
705 self
._print
_sig
_code
(func
, s
)
707 except Exception as e
:
708 tb
= traceback
.format_exc()
709 raise ExtractorError(
710 'Signature extraction failed: ' + tb
, cause
=e
)
712 def _get_subtitles(self
, video_id
, webpage
):
714 subs_doc
= self
._download
_xml
(
715 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
,
716 video_id
, note
=False)
717 except ExtractorError
as err
:
718 self
._downloader
.report_warning('unable to download video subtitles: %s' % compat_str(err
))
722 for track
in subs_doc
.findall('track'):
723 lang
= track
.attrib
['lang_code']
724 if lang
in sub_lang_list
:
727 for ext
in ['sbv', 'vtt', 'srt']:
728 params
= compat_urllib_parse
.urlencode({
732 'name': track
.attrib
['name'].encode('utf-8'),
735 'url': 'https://www.youtube.com/api/timedtext?' + params
,
738 sub_lang_list
[lang
] = sub_formats
739 if not sub_lang_list
:
740 self
._downloader
.report_warning('video doesn\'t have subtitles')
744 def _get_automatic_captions(self
, video_id
, webpage
):
745 """We need the webpage for getting the captions url, pass it as an
746 argument to speed up the process."""
747 self
.to_screen('%s: Looking for automatic captions' % video_id
)
748 mobj
= re
.search(r
';ytplayer.config = ({.*?});', webpage
)
749 err_msg
= 'Couldn\'t find automatic captions for %s' % video_id
751 self
._downloader
.report_warning(err_msg
)
753 player_config
= json
.loads(mobj
.group(1))
755 args
= player_config
['args']
756 caption_url
= args
['ttsurl']
757 timestamp
= args
['timestamp']
758 # We get the available subtitles
759 list_params
= compat_urllib_parse
.urlencode({
764 list_url
= caption_url
+ '&' + list_params
765 caption_list
= self
._download
_xml
(list_url
, video_id
)
766 original_lang_node
= caption_list
.find('track')
767 if original_lang_node
is None:
768 self
._downloader
.report_warning('Video doesn\'t have automatic captions')
770 original_lang
= original_lang_node
.attrib
['lang_code']
771 caption_kind
= original_lang_node
.attrib
.get('kind', '')
774 for lang_node
in caption_list
.findall('target'):
775 sub_lang
= lang_node
.attrib
['lang_code']
777 for ext
in ['sbv', 'vtt', 'srt']:
778 params
= compat_urllib_parse
.urlencode({
779 'lang': original_lang
,
783 'kind': caption_kind
,
786 'url': caption_url
+ '&' + params
,
789 sub_lang_list
[sub_lang
] = sub_formats
791 # An extractor error can be raise by the download process if there are
792 # no automatic captions but there are subtitles
793 except (KeyError, ExtractorError
):
794 self
._downloader
.report_warning(err_msg
)
798 def extract_id(cls
, url
):
799 mobj
= re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
)
801 raise ExtractorError('Invalid URL: %s' % url
)
802 video_id
= mobj
.group(2)
805 def _extract_from_m3u8(self
, manifest_url
, video_id
):
808 def _get_urls(_manifest
):
809 lines
= _manifest
.split('\n')
810 urls
= filter(lambda l
: l
and not l
.startswith('#'),
813 manifest
= self
._download
_webpage
(manifest_url
, video_id
, 'Downloading formats manifest')
814 formats_urls
= _get_urls(manifest
)
815 for format_url
in formats_urls
:
816 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
817 url_map
[itag
] = format_url
820 def _extract_annotations(self
, video_id
):
821 url
= 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
822 return self
._download
_webpage
(url
, video_id
, note
='Searching for annotations.', errnote
='Unable to download video annotations.')
824 def _parse_dash_manifest(
825 self
, video_id
, dash_manifest_url
, player_url
, age_gate
, fatal
=True):
826 def decrypt_sig(mobj
):
828 dec_s
= self
._decrypt
_signature
(s
, video_id
, player_url
, age_gate
)
829 return '/signature/%s' % dec_s
830 dash_manifest_url
= re
.sub(r
'/s/([a-fA-F0-9\.]+)', decrypt_sig
, dash_manifest_url
)
831 dash_doc
= self
._download
_xml
(
832 dash_manifest_url
, video_id
,
833 note
='Downloading DASH manifest',
834 errnote
='Could not download DASH manifest',
837 if dash_doc
is False:
841 for a
in dash_doc
.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
842 mime_type
= a
.attrib
.get('mimeType')
843 for r
in a
.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
844 url_el
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
847 if mime_type
== 'text/vtt':
848 # TODO implement WebVTT downloading
850 elif mime_type
.startswith('audio/') or mime_type
.startswith('video/'):
851 segment_list
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}SegmentList')
852 format_id
= r
.attrib
['id']
853 video_url
= url_el
.text
854 filesize
= int_or_none(url_el
.attrib
.get('{http://youtube.com/yt/2012/10/10}contentLength'))
856 'format_id': format_id
,
858 'width': int_or_none(r
.attrib
.get('width')),
859 'height': int_or_none(r
.attrib
.get('height')),
860 'tbr': int_or_none(r
.attrib
.get('bandwidth'), 1000),
861 'asr': int_or_none(r
.attrib
.get('audioSamplingRate')),
862 'filesize': filesize
,
863 'fps': int_or_none(r
.attrib
.get('frameRate')),
865 if segment_list
is not None:
867 'initialization_url': segment_list
.find('{urn:mpeg:DASH:schema:MPD:2011}Initialization').attrib
['sourceURL'],
868 'segment_urls': [segment
.attrib
.get('media') for segment
in segment_list
.findall('{urn:mpeg:DASH:schema:MPD:2011}SegmentURL')],
869 'protocol': 'http_dash_segments',
872 existing_format
= next(
874 if fo
['format_id'] == format_id
)
875 except StopIteration:
876 full_info
= self
._formats
.get(format_id
, {}).copy()
878 codecs
= r
.attrib
.get('codecs')
880 if full_info
.get('acodec') == 'none' and 'vcodec' not in full_info
:
881 full_info
['vcodec'] = codecs
882 elif full_info
.get('vcodec') == 'none' and 'acodec' not in full_info
:
883 full_info
['acodec'] = codecs
884 formats
.append(full_info
)
886 existing_format
.update(f
)
888 self
.report_warning('Unknown MIME type %s in DASH manifest' % mime_type
)
891 def _real_extract(self
, url
):
893 'http' if self
._downloader
.params
.get('prefer_insecure', False)
898 parsed_url
= compat_urllib_parse_urlparse(url
)
899 for component
in [parsed_url
.fragment
, parsed_url
.query
]:
900 query
= compat_parse_qs(component
)
901 if start_time
is None and 't' in query
:
902 start_time
= parse_duration(query
['t'][0])
903 if start_time
is None and 'start' in query
:
904 start_time
= parse_duration(query
['start'][0])
905 if end_time
is None and 'end' in query
:
906 end_time
= parse_duration(query
['end'][0])
908 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
909 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
911 url
= proto
+ '://www.youtube.com/' + compat_urllib_parse_unquote(mobj
.group(1)).lstrip('/')
912 video_id
= self
.extract_id(url
)
915 url
= proto
+ '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
916 video_webpage
= self
._download
_webpage
(url
, video_id
)
918 # Attempt to extract SWF player URL
919 mobj
= re
.search(r
'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
921 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
927 def add_dash_mpd(video_info
):
928 dash_mpd
= video_info
.get('dashmpd')
929 if dash_mpd
and dash_mpd
[0] not in dash_mpds
:
930 dash_mpds
.append(dash_mpd
[0])
935 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
937 # We simulate the access to the video from www.youtube.com/v/{video_id}
938 # this can be viewed without login into Youtube
939 url
= proto
+ '://www.youtube.com/embed/%s' % video_id
940 embed_webpage
= self
._download
_webpage
(url
, video_id
, 'Downloading embed webpage')
941 data
= compat_urllib_parse
.urlencode({
942 'video_id': video_id
,
943 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
944 'sts': self
._search
_regex
(
945 r
'"sts"\s*:\s*(\d+)', embed_webpage
, 'sts', default
=''),
947 video_info_url
= proto
+ '://www.youtube.com/get_video_info?' + data
948 video_info_webpage
= self
._download
_webpage
(
949 video_info_url
, video_id
,
950 note
='Refetching age-gated info webpage',
951 errnote
='unable to download video info webpage')
952 video_info
= compat_parse_qs(video_info_webpage
)
953 add_dash_mpd(video_info
)
957 # Try looking directly into the video webpage
958 mobj
= re
.search(r
';ytplayer\.config\s*=\s*({.*?});', video_webpage
)
960 json_code
= uppercase_escape(mobj
.group(1))
961 ytplayer_config
= json
.loads(json_code
)
962 args
= ytplayer_config
['args']
963 if args
.get('url_encoded_fmt_stream_map'):
964 # Convert to the same format returned by compat_parse_qs
965 video_info
= dict((k
, [v
]) for k
, v
in args
.items())
966 add_dash_mpd(video_info
)
967 if args
.get('livestream') == '1' or args
.get('live_playback') == 1:
969 if not video_info
or self
._downloader
.params
.get('youtube_include_dash_manifest', True):
970 # We also try looking in get_video_info since it may contain different dashmpd
971 # URL that points to a DASH manifest with possibly different itag set (some itags
972 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
973 # manifest pointed by get_video_info's dashmpd).
974 # The general idea is to take a union of itags of both DASH manifests (for example
975 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
976 self
.report_video_info_webpage_download(video_id
)
977 for el_type
in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
979 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
980 % (proto
, video_id
, el_type
))
981 video_info_webpage
= self
._download
_webpage
(
983 video_id
, note
=False,
984 errnote
='unable to download video info webpage')
985 get_video_info
= compat_parse_qs(video_info_webpage
)
986 if get_video_info
.get('use_cipher_signature') != ['True']:
987 add_dash_mpd(get_video_info
)
989 video_info
= get_video_info
990 if 'token' in get_video_info
:
992 if 'token' not in video_info
:
993 if 'reason' in video_info
:
994 if 'The uploader has not made this video available in your country.' in video_info
['reason']:
995 regions_allowed
= self
._html
_search
_meta
('regionsAllowed', video_webpage
, default
=None)
997 raise ExtractorError('YouTube said: This video is available in %s only' % (
998 ', '.join(map(ISO3166Utils
.short2full
, regions_allowed
.split(',')))),
1000 raise ExtractorError(
1001 'YouTube said: %s' % video_info
['reason'][0],
1002 expected
=True, video_id
=video_id
)
1004 raise ExtractorError(
1005 '"token" parameter not in video info for unknown reason',
1008 if 'view_count' in video_info
:
1009 view_count
= int(video_info
['view_count'][0])
1013 # Check for "rental" videos
1014 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
1015 raise ExtractorError('"rental" videos not supported')
1017 # Start extracting information
1018 self
.report_information_extraction(video_id
)
1021 if 'author' not in video_info
:
1022 raise ExtractorError('Unable to extract uploader name')
1023 video_uploader
= compat_urllib_parse_unquote_plus(video_info
['author'][0])
1026 video_uploader_id
= None
1027 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
1028 if mobj
is not None:
1029 video_uploader_id
= mobj
.group(1)
1031 self
._downloader
.report_warning('unable to extract uploader nickname')
1034 if 'title' in video_info
:
1035 video_title
= video_info
['title'][0]
1037 self
._downloader
.report_warning('Unable to extract video title')
1041 # We try first to get a high quality image:
1042 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
1043 video_webpage
, re
.DOTALL
)
1044 if m_thumb
is not None:
1045 video_thumbnail
= m_thumb
.group(1)
1046 elif 'thumbnail_url' not in video_info
:
1047 self
._downloader
.report_warning('unable to extract video thumbnail')
1048 video_thumbnail
= None
1049 else: # don't panic if we can't find it
1050 video_thumbnail
= compat_urllib_parse_unquote_plus(video_info
['thumbnail_url'][0])
1053 upload_date
= self
._html
_search
_meta
(
1054 'datePublished', video_webpage
, 'upload date', default
=None)
1056 upload_date
= self
._search
_regex
(
1057 [r
'(?s)id="eow-date.*?>(.*?)</span>',
1058 r
'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1059 video_webpage
, 'upload date', default
=None)
1061 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
1062 upload_date
= unified_strdate(upload_date
)
1064 m_cat_container
= self
._search
_regex
(
1065 r
'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1066 video_webpage
, 'categories', default
=None)
1068 category
= self
._html
_search
_regex
(
1069 r
'(?s)<a[^<]+>(.*?)</a>', m_cat_container
, 'category',
1071 video_categories
= None if category
is None else [category
]
1073 video_categories
= None
1076 video_description
= get_element_by_id("eow-description", video_webpage
)
1077 if video_description
:
1078 video_description
= re
.sub(r
'''(?x)
1080 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1082 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1083 class="yt-uix-redirect-link"\s*>
1086 ''', r
'\1', video_description
)
1087 video_description
= clean_html(video_description
)
1089 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
1091 video_description
= unescapeHTML(fd_mobj
.group(1))
1093 video_description
= ''
1095 def _extract_count(count_name
):
1096 return str_to_int(self
._search
_regex
(
1097 r
'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1098 % re
.escape(count_name
),
1099 video_webpage
, count_name
, default
=None))
1101 like_count
= _extract_count('like')
1102 dislike_count
= _extract_count('dislike')
1105 video_subtitles
= self
.extract_subtitles(video_id
, video_webpage
)
1106 automatic_captions
= self
.extract_automatic_captions(video_id
, video_webpage
)
1108 if 'length_seconds' not in video_info
:
1109 self
._downloader
.report_warning('unable to extract video duration')
1110 video_duration
= None
1112 video_duration
= int(compat_urllib_parse_unquote_plus(video_info
['length_seconds'][0]))
1115 video_annotations
= None
1116 if self
._downloader
.params
.get('writeannotations', False):
1117 video_annotations
= self
._extract
_annotations
(video_id
)
1119 def _map_to_format_list(urlmap
):
1121 for itag
, video_real_url
in urlmap
.items():
1124 'url': video_real_url
,
1125 'player_url': player_url
,
1127 if itag
in self
._formats
:
1128 dct
.update(self
._formats
[itag
])
1132 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
1133 self
.report_rtmp_download()
1135 'format_id': '_rtmp',
1137 'url': video_info
['conn'][0],
1138 'player_url': player_url
,
1140 elif len(video_info
.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info
.get('adaptive_fmts', [''])[0]) >= 1:
1141 encoded_url_map
= video_info
.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info
.get('adaptive_fmts', [''])[0]
1142 if 'rtmpe%3Dyes' in encoded_url_map
:
1143 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
1145 for url_data_str
in encoded_url_map
.split(','):
1146 url_data
= compat_parse_qs(url_data_str
)
1147 if 'itag' not in url_data
or 'url' not in url_data
:
1149 format_id
= url_data
['itag'][0]
1150 url
= url_data
['url'][0]
1152 if 'sig' in url_data
:
1153 url
+= '&signature=' + url_data
['sig'][0]
1154 elif 's' in url_data
:
1155 encrypted_sig
= url_data
['s'][0]
1156 ASSETS_RE
= r
'"assets":.+?"js":\s*("[^"]+")'
1158 jsplayer_url_json
= self
._search
_regex
(
1160 embed_webpage
if age_gate
else video_webpage
,
1161 'JS player URL (1)', default
=None)
1162 if not jsplayer_url_json
and not age_gate
:
1163 # We need the embed website after all
1164 if embed_webpage
is None:
1165 embed_url
= proto
+ '://www.youtube.com/embed/%s' % video_id
1166 embed_webpage
= self
._download
_webpage
(
1167 embed_url
, video_id
, 'Downloading embed webpage')
1168 jsplayer_url_json
= self
._search
_regex
(
1169 ASSETS_RE
, embed_webpage
, 'JS player URL')
1171 player_url
= json
.loads(jsplayer_url_json
)
1172 if player_url
is None:
1173 player_url_json
= self
._search
_regex
(
1174 r
'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1175 video_webpage
, 'age gate player URL')
1176 player_url
= json
.loads(player_url_json
)
1178 if self
._downloader
.params
.get('verbose'):
1179 if player_url
is None:
1180 player_version
= 'unknown'
1181 player_desc
= 'unknown'
1183 if player_url
.endswith('swf'):
1184 player_version
= self
._search
_regex
(
1185 r
'-(.+?)(?:/watch_as3)?\.swf$', player_url
,
1186 'flash player', fatal
=False)
1187 player_desc
= 'flash player %s' % player_version
1189 player_version
= self
._search
_regex
(
1190 r
'html5player-([^/]+?)(?:/html5player)?\.js',
1192 'html5 player', fatal
=False)
1193 player_desc
= 'html5 player %s' % player_version
1195 parts_sizes
= self
._signature
_cache
_id
(encrypted_sig
)
1196 self
.to_screen('{%s} signature length %s, %s' %
1197 (format_id
, parts_sizes
, player_desc
))
1199 signature
= self
._decrypt
_signature
(
1200 encrypted_sig
, video_id
, player_url
, age_gate
)
1201 url
+= '&signature=' + signature
1202 if 'ratebypass' not in url
:
1203 url
+= '&ratebypass=yes'
1204 url_map
[format_id
] = url
1205 formats
= _map_to_format_list(url_map
)
1206 elif video_info
.get('hlsvp'):
1207 manifest_url
= video_info
['hlsvp'][0]
1208 url_map
= self
._extract
_from
_m
3u8(manifest_url
, video_id
)
1209 formats
= _map_to_format_list(url_map
)
1211 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1213 # Look for the DASH manifest
1214 if self
._downloader
.params
.get('youtube_include_dash_manifest', True):
1215 dash_mpd_fatal
= True
1216 for dash_manifest_url
in dash_mpds
:
1219 for df
in self
._parse
_dash
_manifest
(
1220 video_id
, dash_manifest_url
, player_url
, age_gate
, dash_mpd_fatal
):
1221 # Do not overwrite DASH format found in some previous DASH manifest
1222 if df
['format_id'] not in dash_formats
:
1223 dash_formats
[df
['format_id']] = df
1224 # Additional DASH manifests may end up in HTTP Error 403 therefore
1225 # allow them to fail without bug report message if we already have
1226 # some DASH manifest succeeded. This is temporary workaround to reduce
1227 # burst of bug reports until we figure out the reason and whether it
1228 # can be fixed at all.
1229 dash_mpd_fatal
= False
1230 except (ExtractorError
, KeyError) as e
:
1231 self
.report_warning(
1232 'Skipping DASH manifest: %r' % e
, video_id
)
1234 # Remove the formats we found through non-DASH, they
1235 # contain less info and it can be wrong, because we use
1236 # fixed values (for example the resolution). See
1237 # https://github.com/rg3/youtube-dl/issues/5774 for an
1239 formats
= [f
for f
in formats
if f
['format_id'] not in dash_formats
.keys()]
1240 formats
.extend(dash_formats
.values())
1242 # Check for malformed aspect ratio
1243 stretched_m
= re
.search(
1244 r
'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1247 ratio
= float(stretched_m
.group('w')) / float(stretched_m
.group('h'))
1249 if f
.get('vcodec') != 'none':
1250 f
['stretched_ratio'] = ratio
1252 self
._sort
_formats
(formats
)
1256 'uploader': video_uploader
,
1257 'uploader_id': video_uploader_id
,
1258 'upload_date': upload_date
,
1259 'title': video_title
,
1260 'thumbnail': video_thumbnail
,
1261 'description': video_description
,
1262 'categories': video_categories
,
1263 'subtitles': video_subtitles
,
1264 'automatic_captions': automatic_captions
,
1265 'duration': video_duration
,
1266 'age_limit': 18 if age_gate
else 0,
1267 'annotations': video_annotations
,
1268 'webpage_url': proto
+ '://www.youtube.com/watch?v=%s' % video_id
,
1269 'view_count': view_count
,
1270 'like_count': like_count
,
1271 'dislike_count': dislike_count
,
1272 'average_rating': float_or_none(video_info
.get('avg_rating', [None])[0]),
1275 'start_time': start_time
,
1276 'end_time': end_time
,
1280 class YoutubePlaylistIE(YoutubeBaseInfoExtractor
):
1281 IE_DESC
= 'YouTube.com playlists'
1282 _VALID_URL
= r
"""(?x)(?:
1287 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1288 \? (?:.*?&)*? (?:p|a|list)=
1292 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1293 # Top tracks, they can also include dots
1298 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1300 _TEMPLATE_URL
= 'https://www.youtube.com/playlist?list=%s'
1301 _VIDEO_RE
= r
'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1302 IE_NAME
= 'youtube:playlist'
1304 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1306 'title': 'ytdl test PL',
1307 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1309 'playlist_count': 3,
1311 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1313 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1314 'title': 'YDL_Empty_List',
1316 'playlist_count': 0,
1318 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1319 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1321 'title': '29C3: Not my department',
1322 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1324 'playlist_count': 95,
1326 'note': 'issue #673',
1327 'url': 'PLBB231211A4F62143',
1329 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1330 'id': 'PLBB231211A4F62143',
1332 'playlist_mincount': 26,
1334 'note': 'Large playlist',
1335 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1337 'title': 'Uploads from Cauchemar',
1338 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1340 'playlist_mincount': 799,
1342 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1344 'title': 'YDL_safe_search',
1345 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1347 'playlist_count': 2,
1350 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1351 'playlist_count': 4,
1354 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1357 'note': 'Embedded SWF player',
1358 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1359 'playlist_count': 4,
1362 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1365 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1366 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1368 'title': 'Uploads from Interstellar Movie',
1369 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1371 'playlist_mincout': 21,
1374 def _real_initialize(self
):
1377 def _extract_mix(self
, playlist_id
):
1378 # The mixes are generated from a single video
1379 # the id of the playlist is just 'RD' + video_id
1380 url
= 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id
[-11:], playlist_id
)
1381 webpage
= self
._download
_webpage
(
1382 url
, playlist_id
, 'Downloading Youtube mix')
1383 search_title
= lambda class_name
: get_element_by_attribute('class', class_name
, webpage
)
1385 search_title('playlist-title') or
1386 search_title('title long-title') or
1387 search_title('title'))
1388 title
= clean_html(title_span
)
1389 ids
= orderedSet(re
.findall(
1390 r
'''(?xs)data-video-username=".*?".*?
1391 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re
.escape(playlist_id
),
1393 url_results
= self
._ids
_to
_results
(ids
)
1395 return self
.playlist_result(url_results
, playlist_id
, title
)
1397 def _extract_playlist(self
, playlist_id
):
1398 url
= self
._TEMPLATE
_URL
% playlist_id
1399 page
= self
._download
_webpage
(url
, playlist_id
)
1401 for match
in re
.findall(r
'<div class="yt-alert-message">([^<]+)</div>', page
):
1402 match
= match
.strip()
1403 # Check if the playlist exists or is private
1404 if re
.match(r
'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match
):
1405 raise ExtractorError(
1406 'The playlist doesn\'t exist or is private, use --username or '
1407 '--netrc to access it.',
1409 elif re
.match(r
'[^<]*Invalid parameters[^<]*', match
):
1410 raise ExtractorError(
1411 'Invalid parameters. Maybe URL is incorrect.',
1413 elif re
.match(r
'[^<]*Choose your language[^<]*', match
):
1416 self
.report_warning('Youtube gives an alert message: ' + match
)
1418 # Extract the video ids from the playlist pages
1420 more_widget_html
= content_html
= page
1421 for page_num
in itertools
.count(1):
1422 matches
= re
.finditer(self
._VIDEO
_RE
, content_html
)
1423 # We remove the duplicates and the link with index 0
1424 # (it's not the first video of the playlist)
1425 new_ids
= orderedSet(m
.group('id') for m
in matches
if m
.group('index') != '0')
1426 for vid_id
in new_ids
:
1427 yield self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
1429 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1433 more
= self
._download
_json
(
1434 'https://youtube.com/%s' % mobj
.group('more'), playlist_id
,
1435 'Downloading page #%s' % page_num
,
1436 transform_source
=uppercase_escape
)
1437 content_html
= more
['content_html']
1438 if not content_html
.strip():
1439 # Some webpages show a "Load more" button but they don't
1442 more_widget_html
= more
['load_more_widget_html']
1444 playlist_title
= self
._html
_search
_regex
(
1445 r
'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1448 return self
.playlist_result(_entries(), playlist_id
, playlist_title
)
1450 def _real_extract(self
, url
):
1451 # Extract playlist id
1452 mobj
= re
.match(self
._VALID
_URL
, url
)
1454 raise ExtractorError('Invalid URL: %s' % url
)
1455 playlist_id
= mobj
.group(1) or mobj
.group(2)
1457 # Check if it's a video-specific URL
1458 query_dict
= compat_urlparse
.parse_qs(compat_urlparse
.urlparse(url
).query
)
1459 if 'v' in query_dict
:
1460 video_id
= query_dict
['v'][0]
1461 if self
._downloader
.params
.get('noplaylist'):
1462 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1463 return self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1465 self
.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id
, video_id
))
1467 if playlist_id
.startswith('RD') or playlist_id
.startswith('UL'):
1468 # Mixes require a custom extraction process
1469 return self
._extract
_mix
(playlist_id
)
1471 return self
._extract
_playlist
(playlist_id
)
1474 class YoutubeChannelIE(InfoExtractor
):
1475 IE_DESC
= 'YouTube.com channels'
1476 _VALID_URL
= r
'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1477 _TEMPLATE_URL
= 'https://www.youtube.com/channel/%s/videos'
1478 IE_NAME
= 'youtube:channel'
1480 'note': 'paginated channel',
1481 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1482 'playlist_mincount': 91,
1484 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1489 def extract_videos_from_page(page
):
1492 for mobj
in re
.finditer(r
'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page
):
1493 video_id
= mobj
.group('id')
1494 video_title
= unescapeHTML(mobj
.group('title'))
1496 idx
= ids_in_page
.index(video_id
)
1497 if video_title
and not titles_in_page
[idx
]:
1498 titles_in_page
[idx
] = video_title
1500 ids_in_page
.append(video_id
)
1501 titles_in_page
.append(video_title
)
1502 return zip(ids_in_page
, titles_in_page
)
1504 def _real_extract(self
, url
):
1505 channel_id
= self
._match
_id
(url
)
1507 url
= self
._TEMPLATE
_URL
% channel_id
1509 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1510 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1511 # otherwise fallback on channel by page extraction
1512 channel_page
= self
._download
_webpage
(
1513 url
+ '?view=57', channel_id
,
1514 'Downloading channel page', fatal
=False)
1515 channel_playlist_id
= self
._html
_search
_meta
(
1516 'channelId', channel_page
, 'channel id', default
=None)
1517 if not channel_playlist_id
:
1518 channel_playlist_id
= self
._search
_regex
(
1519 r
'data-channel-external-id="([^"]+)"',
1520 channel_page
, 'channel id', default
=None)
1521 if channel_playlist_id
and channel_playlist_id
.startswith('UC'):
1522 playlist_id
= 'UU' + channel_playlist_id
[2:]
1523 return self
.url_result(
1524 compat_urlparse
.urljoin(url
, '/playlist?list=%s' % playlist_id
), 'YoutubePlaylist')
1526 channel_page
= self
._download
_webpage
(url
, channel_id
, 'Downloading page #1')
1527 autogenerated
= re
.search(r
'''(?x)
1529 channel-header-autogenerated-label|
1530 yt-channel-title-autogenerated
1531 )[^"]*"''', channel_page
) is not None
1534 # The videos are contained in a single page
1535 # the ajax pages can't be used, they are empty
1538 video_id
, 'Youtube', video_id
=video_id
,
1539 video_title
=video_title
)
1540 for video_id
, video_title
in self
.extract_videos_from_page(channel_page
)]
1541 return self
.playlist_result(entries
, channel_id
)
1544 more_widget_html
= content_html
= channel_page
1545 for pagenum
in itertools
.count(1):
1547 for video_id
, video_title
in self
.extract_videos_from_page(content_html
):
1548 yield self
.url_result(
1549 video_id
, 'Youtube', video_id
=video_id
,
1550 video_title
=video_title
)
1553 r
'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1558 more
= self
._download
_json
(
1559 'https://youtube.com/%s' % mobj
.group('more'), channel_id
,
1560 'Downloading page #%s' % (pagenum
+ 1),
1561 transform_source
=uppercase_escape
)
1562 content_html
= more
['content_html']
1563 more_widget_html
= more
['load_more_widget_html']
1565 return self
.playlist_result(_entries(), channel_id
)
1568 class YoutubeUserIE(YoutubeChannelIE
):
1569 IE_DESC
= 'YouTube.com user videos (URL or "ytuser" keyword)'
1570 _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_-]+)'
1571 _TEMPLATE_URL
= 'https://www.youtube.com/user/%s/videos'
1572 IE_NAME
= 'youtube:user'
1575 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1576 'playlist_mincount': 320,
1578 'title': 'TheLinuxFoundation',
1581 'url': 'ytuser:phihag',
1582 'only_matching': True,
1586 def suitable(cls
, url
):
1587 # Don't return True if the url can be extracted with other youtube
1588 # extractor, the regex would is too permissive and it would match.
1589 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1590 if any(ie
.suitable(url
) for ie
in other_ies
):
1593 return super(YoutubeUserIE
, cls
).suitable(url
)
1596 class YoutubeSearchIE(SearchInfoExtractor
, YoutubePlaylistIE
):
1597 IE_DESC
= 'YouTube.com searches'
1598 # there doesn't appear to be a real limit, for example if you search for
1599 # 'python' you get more than 8.000.000 results
1600 _MAX_RESULTS
= float('inf')
1601 IE_NAME
= 'youtube:search'
1602 _SEARCH_KEY
= 'ytsearch'
1603 _EXTRA_QUERY_ARGS
= {}
1606 def _get_n_results(self
, query
, n
):
1607 """Get a specified number of results for a query"""
1612 for pagenum
in itertools
.count(1):
1614 'search_query': query
.encode('utf-8'),
1618 url_query
.update(self
._EXTRA
_QUERY
_ARGS
)
1619 result_url
= 'https://www.youtube.com/results?' + compat_urllib_parse
.urlencode(url_query
)
1620 data
= self
._download
_json
(
1621 result_url
, video_id
='query "%s"' % query
,
1622 note
='Downloading page %s' % pagenum
,
1623 errnote
='Unable to download API page')
1624 html_content
= data
[1]['body']['content']
1626 if 'class="search-message' in html_content
:
1627 raise ExtractorError(
1628 '[youtube] No video results', expected
=True)
1630 new_videos
= self
._ids
_to
_results
(orderedSet(re
.findall(
1631 r
'href="/watch\?v=(.{11})', html_content
)))
1632 videos
+= new_videos
1633 if not new_videos
or len(videos
) > limit
:
1638 return self
.playlist_result(videos
, query
)
1641 class YoutubeSearchDateIE(YoutubeSearchIE
):
1642 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1643 _SEARCH_KEY
= 'ytsearchdate'
1644 IE_DESC
= 'YouTube.com searches, newest videos first'
1645 _EXTRA_QUERY_ARGS
= {'search_sort': 'video_date_uploaded'}
1648 class YoutubeSearchURLIE(InfoExtractor
):
1649 IE_DESC
= 'YouTube.com search URLs'
1650 IE_NAME
= 'youtube:search_url'
1651 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1653 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1654 'playlist_mincount': 5,
1656 'title': 'youtube-dl test video',
1660 def _real_extract(self
, url
):
1661 mobj
= re
.match(self
._VALID
_URL
, url
)
1662 query
= compat_urllib_parse_unquote_plus(mobj
.group('query'))
1664 webpage
= self
._download
_webpage
(url
, query
)
1665 result_code
= self
._search
_regex
(
1666 r
'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage
, 'result HTML')
1668 part_codes
= re
.findall(
1669 r
'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code
)
1671 for part_code
in part_codes
:
1672 part_title
= self
._html
_search
_regex
(
1673 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1674 part_url_snippet
= self
._html
_search
_regex
(
1675 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1676 part_url
= compat_urlparse
.urljoin(
1677 'https://www.youtube.com/', part_url_snippet
)
1681 'title': part_title
,
1685 '_type': 'playlist',
1691 class YoutubeShowIE(InfoExtractor
):
1692 IE_DESC
= 'YouTube.com (multi-season) shows'
1693 _VALID_URL
= r
'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1694 IE_NAME
= 'youtube:show'
1696 'url': 'http://www.youtube.com/show/airdisasters',
1697 'playlist_mincount': 3,
1699 'id': 'airdisasters',
1700 'title': 'Air Disasters',
1704 def _real_extract(self
, url
):
1705 mobj
= re
.match(self
._VALID
_URL
, url
)
1706 playlist_id
= mobj
.group('id')
1707 webpage
= self
._download
_webpage
(
1708 url
, playlist_id
, 'Downloading show webpage')
1709 # There's one playlist for each season of the show
1710 m_seasons
= list(re
.finditer(r
'href="(/playlist\?list=.*?)"', webpage
))
1711 self
.to_screen('%s: Found %s seasons' % (playlist_id
, len(m_seasons
)))
1714 'https://www.youtube.com' + season
.group(1), 'YoutubePlaylist')
1715 for season
in m_seasons
1717 title
= self
._og
_search
_title
(webpage
, fatal
=False)
1720 '_type': 'playlist',
1727 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1729 Base class for feed extractors
1730 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1732 _LOGIN_REQUIRED
= True
1736 return 'youtube:%s' % self
._FEED
_NAME
1738 def _real_initialize(self
):
1741 def _real_extract(self
, url
):
1742 page
= self
._download
_webpage
(
1743 'https://www.youtube.com/feed/%s' % self
._FEED
_NAME
, self
._PLAYLIST
_TITLE
)
1745 # The extraction process is the same as for playlists, but the regex
1746 # for the video ids doesn't contain an index
1748 more_widget_html
= content_html
= page
1749 for page_num
in itertools
.count(1):
1750 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
1752 # 'recommended' feed has infinite 'load more' and each new portion spins
1753 # the same videos in (sometimes) slightly different order, so we'll check
1754 # for unicity and break when portion has no new videos
1755 new_ids
= filter(lambda video_id
: video_id
not in ids
, orderedSet(matches
))
1761 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1765 more
= self
._download
_json
(
1766 'https://youtube.com/%s' % mobj
.group('more'), self
._PLAYLIST
_TITLE
,
1767 'Downloading page #%s' % page_num
,
1768 transform_source
=uppercase_escape
)
1769 content_html
= more
['content_html']
1770 more_widget_html
= more
['load_more_widget_html']
1772 return self
.playlist_result(
1773 self
._ids
_to
_results
(ids
), playlist_title
=self
._PLAYLIST
_TITLE
)
1776 class YoutubeWatchLaterIE(YoutubePlaylistIE
):
1777 IE_NAME
= 'youtube:watchlater'
1778 IE_DESC
= 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1779 _VALID_URL
= r
'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1781 _TESTS
= [] # override PlaylistIE tests
1783 def _real_extract(self
, url
):
1784 return self
._extract
_playlist
('WL')
1787 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
1788 IE_NAME
= 'youtube:favorites'
1789 IE_DESC
= 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1790 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1791 _LOGIN_REQUIRED
= True
1793 def _real_extract(self
, url
):
1794 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1795 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, 'favourites playlist id')
1796 return self
.url_result(playlist_id
, 'YoutubePlaylist')
1799 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
1800 IE_DESC
= 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1801 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1802 _FEED_NAME
= 'recommended'
1803 _PLAYLIST_TITLE
= 'Youtube Recommended videos'
1806 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor
):
1807 IE_DESC
= 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1808 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1809 _FEED_NAME
= 'subscriptions'
1810 _PLAYLIST_TITLE
= 'Youtube Subscriptions'
1813 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
1814 IE_DESC
= 'Youtube watch history, ":ythistory" for short (requires authentication)'
1815 _VALID_URL
= 'https?://www\.youtube\.com/feed/history|:ythistory'
1816 _FEED_NAME
= 'history'
1817 _PLAYLIST_TITLE
= 'Youtube History'
1820 class YoutubeTruncatedURLIE(InfoExtractor
):
1821 IE_NAME
= 'youtube:truncated_url'
1822 IE_DESC
= False # Do not list
1823 _VALID_URL
= r
'''(?x)
1825 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1828 annotation_id=annotation_[^&]+|
1833 attribution_link\?a=[^&]+
1839 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1840 'only_matching': True,
1842 'url': 'http://www.youtube.com/watch?',
1843 'only_matching': True,
1845 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1846 'only_matching': True,
1848 'url': 'https://www.youtube.com/watch?feature=foo',
1849 'only_matching': True,
1851 'url': 'https://www.youtube.com/watch?hl=en-GB',
1852 'only_matching': True,
1855 def _real_extract(self
, url
):
1856 raise ExtractorError(
1857 'Did you forget to quote the URL? Remember that & is a meta '
1858 'character in most shells, so you want to put the URL in quotes, '
1860 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1861 ' or simply youtube-dl BaW_jenozKc .',
1865 class YoutubeTruncatedIDIE(InfoExtractor
):
1866 IE_NAME
= 'youtube:truncated_id'
1867 IE_DESC
= False # Do not list
1868 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1871 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1872 'only_matching': True,
1875 def _real_extract(self
, url
):
1876 video_id
= self
._match
_id
(url
)
1877 raise ExtractorError(
1878 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id
, url
),