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_request
,
30 get_element_by_attribute
,
42 class YoutubeBaseInfoExtractor(InfoExtractor
):
43 """Provide base functions for Youtube extractors"""
44 _LOGIN_URL
= 'https://accounts.google.com/ServiceLogin'
45 _TWOFACTOR_URL
= 'https://accounts.google.com/SecondFactor'
46 _NETRC_MACHINE
= 'youtube'
47 # If True it will raise an error if no login info is provided
48 _LOGIN_REQUIRED
= False
50 def _set_language(self
):
52 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
53 # YouTube sets the expire time to about two months
54 expire_time
=time
.time() + 2 * 30 * 24 * 3600)
56 def _ids_to_results(self
, ids
):
58 self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
63 Attempt to log in to YouTube.
64 True is returned if successful or skipped.
65 False is returned if login failed.
67 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
69 (username
, password
) = self
._get
_login
_info
()
70 # No authentication to be performed
72 if self
._LOGIN
_REQUIRED
:
73 raise ExtractorError('No login info available, needed for using %s.' % self
.IE_NAME
, expected
=True)
76 login_page
= self
._download
_webpage
(
77 self
._LOGIN
_URL
, None,
78 note
='Downloading login page',
79 errnote
='unable to fetch login page', fatal
=False)
80 if login_page
is False:
83 galx
= self
._search
_regex
(r
'(?s)<input.+?name="GALX".+?value="(.+?)"',
84 login_page
, 'Login GALX parameter')
88 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
93 'PersistentCookie': 'yes',
95 'bgresponse': 'js_disabled',
96 'checkConnection': '',
97 'checkedDomains': 'youtube',
104 'service': 'youtube',
109 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
111 login_form
= dict((k
.encode('utf-8'), v
.encode('utf-8')) for k
, v
in login_form_strs
.items())
112 login_data
= compat_urllib_parse
.urlencode(login_form
).encode('ascii')
114 req
= compat_urllib_request
.Request(self
._LOGIN
_URL
, login_data
)
115 login_results
= self
._download
_webpage
(
117 note
='Logging in', errnote
='unable to log in', fatal
=False)
118 if login_results
is False:
121 if re
.search(r
'id="errormsg_0_Passwd"', login_results
) is not None:
122 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected
=True)
125 # TODO add SMS and phone call support - these require making a request and then prompting the user
127 if re
.search(r
'(?i)<form[^>]* id="gaia_secondfactorform"', login_results
) is not None:
128 tfa_code
= self
._get
_tfa
_info
()
131 self
._downloader
.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
132 self
._downloader
.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
135 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
137 match
= re
.search(r
'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
139 self._downloader.report_warning('Failed to get secTok
- did the page structure change?
')
140 secTok = match.group(1)
141 match = re.search(r'id="timeStmp"\n\s
+value
=\'(.+)\'/>', login_results, re.M | re.U)
143 self._downloader.report_warning('Failed to get timeStmp
- did the page structure change?
')
144 timeStmp = match.group(1)
147 'continue': 'https
://www
.youtube
.com
/signin?action_handle_signin
=true
&feature
=sign_in_button
&hl
=en_US
&nomobiletemp
=1',
149 'smsUserPin
': tfa_code,
150 'smsVerifyPin
': 'Verify
',
152 'PersistentCookie
': 'yes
',
153 'checkConnection
': '',
154 'checkedDomains
': 'youtube
',
157 'timeStmp
': timeStmp,
158 'service
': 'youtube
',
161 tfa_form = dict((k.encode('utf
-8'), v.encode('utf
-8')) for k, v in tfa_form_strs.items())
162 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii
')
164 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
165 tfa_results = self._download_webpage(
167 note='Submitting TFA code
', errnote='unable to submit tfa
', fatal=False)
169 if tfa_results is False:
172 if re.search(r'(?i
)<form
[^
>]* id="gaia_secondfactorform"', tfa_results) is not None:
173 self._downloader.report_warning('Two
-factor code expired
. Please
try again
, or use a one
-use backup code instead
.')
175 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', tfa_results) is not None:
176 self._downloader.report_warning('unable to log
in - did the page structure change?
')
178 if re.search(r'smsauth
-interstitial
-reviewsettings
', tfa_results) is not None:
179 self._downloader.report_warning('Your Google account has a security notice
. Please log
in on your web browser
, resolve the notice
, and try again
.')
182 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', login_results) is not None:
183 self._downloader.report_warning('unable to log
in: bad username
or password
')
187 def _real_initialize(self):
188 if self._downloader is None:
191 if not self._login():
195 class YoutubeIE(YoutubeBaseInfoExtractor):
196 IE_DESC = 'YouTube
.com
'
197 _VALID_URL = r"""(?x)^
199 (?:https?://|//) # http(s):// or protocol-independent URL
200 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
201 (?:www\.)?deturl\.com/www\.youtube\.com/|
202 (?:www\.)?pwnyoutube\.com/|
203 (?:www\.)?yourepeat\.com/|
204 tube\.majestyc\.net/|
205 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
206 (?:.*?\#/)? # handle anchor (#/) redirect urls
207 (?: # the various things that can precede the ID:
208 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
209 |(?: # or the v= param in all its forms
210 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
211 (?:\?|\#!?) # the params delimiter ? or # or #!
212 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
216 |youtu\.be/ # just youtu.be/xxxx
217 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
219 )? # all until now is optional -> you can pass the naked ID
220 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
221 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
222 (?(1).+)? # if we found the ID, everything can follow
224 _NEXT_URL_RE = r'[\?&]next_url
=([^
&]+)'
226 '5': {'ext': 'flv', 'width': 400, 'height': 240},
227 '6': {'ext': 'flv', 'width': 450, 'height': 270},
228 '13': {'ext': '3gp'},
229 '17': {'ext': '3gp', 'width': 176, 'height': 144},
230 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
231 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
232 '34': {'ext': 'flv', 'width': 640, 'height': 360},
233 '35': {'ext': 'flv', 'width': 854, 'height': 480},
234 '36': {'ext': '3gp', 'width': 320, 'height': 240},
235 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
236 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
237 '43': {'ext': 'webm', 'width': 640, 'height': 360},
238 '44': {'ext': 'webm', 'width': 854, 'height': 480},
239 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
240 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
241 '59': {'ext': 'mp4', 'width': 854, 'height': 480},
242 '78': {'ext': 'mp4', 'width': 854, 'height': 480},
246 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
247 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
248 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
249 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
250 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
251 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
252 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
254 # Apple HTTP Live Streaming
255 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
256 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
257 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
258 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
259 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
260 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
261 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
264 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
265 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
266 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
267 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
268 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
269 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
270 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
271 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
272 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
273 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
274 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
277 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
278 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
279 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
282 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
283 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
284 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
285 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
286 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
287 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
288 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
289 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
290 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
291 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
293 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
298 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
299 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
300 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
301 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
302 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
305 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
306 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
308 # Dash webm audio with opus inside
309 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
310 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
311 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
314 '_rtmp
': {'protocol': 'rtmp'},
320 'url
': 'http
://www
.youtube
.com
/watch?v
=BaW_jenozKc
',
324 'title
': 'youtube
-dl test video
"\'/\\ä↭𝕐',
325 'uploader': 'Philipp Hagemeister',
326 'uploader_id': 'phihag',
327 'upload_date': '20121002',
328 '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 .',
329 'categories
': ['Science
& Technology
'],
331 'dislike_count
': int,
335 'url
': 'http
://www
.youtube
.com
/watch?v
=UxxajLWwzqY
',
336 'note
': 'Test generic use_cipher_signature
video (#897)',
340 'upload_date': '20120506',
341 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
342 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
343 'uploader': 'Icona Pop',
344 'uploader_id': 'IconaPop',
348 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
349 'note': 'Test VEVO video with age protection (#956)',
353 'upload_date': '20130703',
354 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
355 'description': 'md5:64249768eec3bc4276236606ea996373',
356 'uploader': 'justintimberlakeVEVO',
357 'uploader_id': 'justintimberlakeVEVO',
361 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
362 'note': 'Embed-only video (#1746)',
366 'upload_date': '20120608',
367 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
368 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
369 'uploader': 'SET India',
370 'uploader_id': 'setindia'
374 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
375 'note': '256k DASH audio (format 141) via DASH manifest',
379 'upload_date': '20121002',
380 'uploader_id': '8KVIDEO',
382 'uploader': '8KVIDEO',
383 'title': 'UHDTV TEST 8K VIDEO.mp4'
386 'youtube_include_dash_manifest': True,
390 # DASH manifest with encrypted signature
392 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
396 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
397 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
398 'uploader': 'AfrojackVEVO',
399 'uploader_id': 'AfrojackVEVO',
400 'upload_date': '20131011',
403 'youtube_include_dash_manifest': True,
407 # JS player signature function name containing $
409 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
413 'title': 'Taylor Swift - Shake It Off',
414 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
415 'uploader': 'TaylorSwiftVEVO',
416 'uploader_id': 'TaylorSwiftVEVO',
417 'upload_date': '20140818',
420 'youtube_include_dash_manifest': True,
426 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
430 'upload_date': '20100909',
431 'uploader': 'The Amazing Atheist',
432 'uploader_id': 'TheAmazingAtheist',
433 'title': 'Burning Everyone\'s Koran',
434 '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',
437 # Normal age-gate video (No vevo, embed allowed)
439 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
443 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
444 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
445 'uploader': 'The Witcher',
446 'uploader_id': 'WitcherGame',
447 'upload_date': '20140605',
450 # Age-gate video with encrypted signature
452 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
456 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
457 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
458 'uploader': 'LloydVEVO',
459 'uploader_id': 'LloydVEVO',
460 'upload_date': '20110629',
463 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
465 'url': '__2ABJjxzNo',
469 'upload_date': '20100430',
470 'uploader_id': 'deadmau5',
471 'description': 'md5:12c56784b8032162bb936a5f76d55360',
472 'uploader': 'deadmau5',
473 'title': 'Deadmau5 - Some Chords (HD)',
475 'expected_warnings': [
476 'DASH manifest missing',
479 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
481 'url': 'lqQg6PlCWgI',
485 'upload_date': '20120731',
486 'uploader_id': 'olympic',
487 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
488 'uploader': 'Olympics',
489 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
492 'skip_download': 'requires avconv',
497 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
501 'stretched_ratio': 16 / 9.,
502 'upload_date': '20110310',
503 'uploader_id': 'AllenMeow',
504 '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',
506 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
509 # url_encoded_fmt_stream_map is empty string
511 'url': 'qEJwOuvDf7I',
515 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
517 'upload_date': '20150404',
518 'uploader_id': 'spbelect',
519 'uploader': 'Наблюдатели Петербурга',
522 'skip_download': 'requires avconv',
525 # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
527 'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
531 'title': 'md5:7b81415841e02ecd4313668cde88737a',
532 'description': 'md5:116377fd2963b81ec4ce64b542173306',
533 'upload_date': '20150625',
534 'uploader_id': 'dorappi2000',
535 'uploader': 'dorappi2000',
536 'formats': 'mincount:33',
541 def __init__(self
, *args
, **kwargs
):
542 super(YoutubeIE
, self
).__init
__(*args
, **kwargs
)
543 self
._player
_cache
= {}
545 def report_video_info_webpage_download(self
, video_id
):
546 """Report attempt to download video info webpage."""
547 self
.to_screen('%s: Downloading video info webpage' % video_id
)
549 def report_information_extraction(self
, video_id
):
550 """Report attempt to extract video information."""
551 self
.to_screen('%s: Extracting video information' % video_id
)
553 def report_unavailable_format(self
, video_id
, format
):
554 """Report extracted video URL."""
555 self
.to_screen('%s: Format %s not available' % (video_id
, format
))
557 def report_rtmp_download(self
):
558 """Indicate the download will use the RTMP protocol."""
559 self
.to_screen('RTMP download detected')
561 def _signature_cache_id(self
, example_sig
):
562 """ Return a string representation of a signature """
563 return '.'.join(compat_str(len(part
)) for part
in example_sig
.split('.'))
565 def _extract_signature_function(self
, video_id
, player_url
, example_sig
):
567 r
'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
570 raise ExtractorError('Cannot identify player %r' % player_url
)
571 player_type
= id_m
.group('ext')
572 player_id
= id_m
.group('id')
574 # Read from filesystem cache
575 func_id
= '%s_%s_%s' % (
576 player_type
, player_id
, self
._signature
_cache
_id
(example_sig
))
577 assert os
.path
.basename(func_id
) == func_id
579 cache_spec
= self
._downloader
.cache
.load('youtube-sigfuncs', func_id
)
580 if cache_spec
is not None:
581 return lambda s
: ''.join(s
[i
] for i
in cache_spec
)
584 'Downloading player %s' % player_url
585 if self
._downloader
.params
.get('verbose') else
586 'Downloading %s player %s' % (player_type
, player_id
)
588 if player_type
== 'js':
589 code
= self
._download
_webpage
(
590 player_url
, video_id
,
592 errnote
='Download of %s failed' % player_url
)
593 res
= self
._parse
_sig
_js
(code
)
594 elif player_type
== 'swf':
595 urlh
= self
._request
_webpage
(
596 player_url
, video_id
,
598 errnote
='Download of %s failed' % player_url
)
600 res
= self
._parse
_sig
_swf
(code
)
602 assert False, 'Invalid player type %r' % player_type
604 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
605 cache_res
= res(test_string
)
606 cache_spec
= [ord(c
) for c
in cache_res
]
608 self
._downloader
.cache
.store('youtube-sigfuncs', func_id
, cache_spec
)
611 def _print_sig_code(self
, func
, example_sig
):
612 def gen_sig_code(idxs
):
613 def _genslice(start
, end
, step
):
614 starts
= '' if start
== 0 else str(start
)
615 ends
= (':%d' % (end
+ step
)) if end
+ step
>= 0 else ':'
616 steps
= '' if step
== 1 else (':%d' % step
)
617 return 's[%s%s%s]' % (starts
, ends
, steps
)
620 # Quelch pyflakes warnings - start will be set when step is set
621 start
= '(Never used)'
622 for i
, prev
in zip(idxs
[1:], idxs
[:-1]):
626 yield _genslice(start
, prev
, step
)
629 if i
- prev
in [-1, 1]:
638 yield _genslice(start
, i
, step
)
640 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
641 cache_res
= func(test_string
)
642 cache_spec
= [ord(c
) for c
in cache_res
]
643 expr_code
= ' + '.join(gen_sig_code(cache_spec
))
644 signature_id_tuple
= '(%s)' % (
645 ', '.join(compat_str(len(p
)) for p
in example_sig
.split('.')))
646 code
= ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
647 ' return %s\n') % (signature_id_tuple
, expr_code
)
648 self
.to_screen('Extracted signature function:\n' + code
)
650 def _parse_sig_js(self
, jscode
):
651 funcname
= self
._search
_regex
(
652 r
'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode
,
653 'Initial JS player signature function name')
655 jsi
= JSInterpreter(jscode
)
656 initial_function
= jsi
.extract_function(funcname
)
657 return lambda s
: initial_function([s
])
659 def _parse_sig_swf(self
, file_contents
):
660 swfi
= SWFInterpreter(file_contents
)
661 TARGET_CLASSNAME
= 'SignatureDecipher'
662 searched_class
= swfi
.extract_class(TARGET_CLASSNAME
)
663 initial_function
= swfi
.extract_function(searched_class
, 'decipher')
664 return lambda s
: initial_function([s
])
666 def _decrypt_signature(self
, s
, video_id
, player_url
, age_gate
=False):
667 """Turn the encrypted s field into a working signature"""
669 if player_url
is None:
670 raise ExtractorError('Cannot decrypt signature without player_url')
672 if player_url
.startswith('//'):
673 player_url
= 'https:' + player_url
675 player_id
= (player_url
, self
._signature
_cache
_id
(s
))
676 if player_id
not in self
._player
_cache
:
677 func
= self
._extract
_signature
_function
(
678 video_id
, player_url
, s
680 self
._player
_cache
[player_id
] = func
681 func
= self
._player
_cache
[player_id
]
682 if self
._downloader
.params
.get('youtube_print_sig_code'):
683 self
._print
_sig
_code
(func
, s
)
685 except Exception as e
:
686 tb
= traceback
.format_exc()
687 raise ExtractorError(
688 'Signature extraction failed: ' + tb
, cause
=e
)
690 def _get_subtitles(self
, video_id
, webpage
):
692 subs_doc
= self
._download
_xml
(
693 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
,
694 video_id
, note
=False)
695 except ExtractorError
as err
:
696 self
._downloader
.report_warning('unable to download video subtitles: %s' % compat_str(err
))
700 for track
in subs_doc
.findall('track'):
701 lang
= track
.attrib
['lang_code']
702 if lang
in sub_lang_list
:
705 for ext
in ['sbv', 'vtt', 'srt']:
706 params
= compat_urllib_parse
.urlencode({
710 'name': track
.attrib
['name'].encode('utf-8'),
713 'url': 'https://www.youtube.com/api/timedtext?' + params
,
716 sub_lang_list
[lang
] = sub_formats
717 if not sub_lang_list
:
718 self
._downloader
.report_warning('video doesn\'t have subtitles')
722 def _get_automatic_captions(self
, video_id
, webpage
):
723 """We need the webpage for getting the captions url, pass it as an
724 argument to speed up the process."""
725 self
.to_screen('%s: Looking for automatic captions' % video_id
)
726 mobj
= re
.search(r
';ytplayer.config = ({.*?});', webpage
)
727 err_msg
= 'Couldn\'t find automatic captions for %s' % video_id
729 self
._downloader
.report_warning(err_msg
)
731 player_config
= json
.loads(mobj
.group(1))
733 args
= player_config
['args']
734 caption_url
= args
['ttsurl']
735 timestamp
= args
['timestamp']
736 # We get the available subtitles
737 list_params
= compat_urllib_parse
.urlencode({
742 list_url
= caption_url
+ '&' + list_params
743 caption_list
= self
._download
_xml
(list_url
, video_id
)
744 original_lang_node
= caption_list
.find('track')
745 if original_lang_node
is None:
746 self
._downloader
.report_warning('Video doesn\'t have automatic captions')
748 original_lang
= original_lang_node
.attrib
['lang_code']
749 caption_kind
= original_lang_node
.attrib
.get('kind', '')
752 for lang_node
in caption_list
.findall('target'):
753 sub_lang
= lang_node
.attrib
['lang_code']
755 for ext
in ['sbv', 'vtt', 'srt']:
756 params
= compat_urllib_parse
.urlencode({
757 'lang': original_lang
,
761 'kind': caption_kind
,
764 'url': caption_url
+ '&' + params
,
767 sub_lang_list
[sub_lang
] = sub_formats
769 # An extractor error can be raise by the download process if there are
770 # no automatic captions but there are subtitles
771 except (KeyError, ExtractorError
):
772 self
._downloader
.report_warning(err_msg
)
776 def extract_id(cls
, url
):
777 mobj
= re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
)
779 raise ExtractorError('Invalid URL: %s' % url
)
780 video_id
= mobj
.group(2)
783 def _extract_from_m3u8(self
, manifest_url
, video_id
):
786 def _get_urls(_manifest
):
787 lines
= _manifest
.split('\n')
788 urls
= filter(lambda l
: l
and not l
.startswith('#'),
791 manifest
= self
._download
_webpage
(manifest_url
, video_id
, 'Downloading formats manifest')
792 formats_urls
= _get_urls(manifest
)
793 for format_url
in formats_urls
:
794 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
795 url_map
[itag
] = format_url
798 def _extract_annotations(self
, video_id
):
799 url
= 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
800 return self
._download
_webpage
(url
, video_id
, note
='Searching for annotations.', errnote
='Unable to download video annotations.')
802 def _parse_dash_manifest(
803 self
, video_id
, dash_manifest_url
, player_url
, age_gate
, fatal
=True):
804 def decrypt_sig(mobj
):
806 dec_s
= self
._decrypt
_signature
(s
, video_id
, player_url
, age_gate
)
807 return '/signature/%s' % dec_s
808 dash_manifest_url
= re
.sub(r
'/s/([a-fA-F0-9\.]+)', decrypt_sig
, dash_manifest_url
)
809 dash_doc
= self
._download
_xml
(
810 dash_manifest_url
, video_id
,
811 note
='Downloading DASH manifest',
812 errnote
='Could not download DASH manifest',
815 if dash_doc
is False:
819 for a
in dash_doc
.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
820 mime_type
= a
.attrib
.get('mimeType')
821 for r
in a
.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
822 url_el
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
825 if mime_type
== 'text/vtt':
826 # TODO implement WebVTT downloading
828 elif mime_type
.startswith('audio/') or mime_type
.startswith('video/'):
829 format_id
= r
.attrib
['id']
830 video_url
= url_el
.text
831 filesize
= int_or_none(url_el
.attrib
.get('{http://youtube.com/yt/2012/10/10}contentLength'))
833 'format_id': format_id
,
835 'width': int_or_none(r
.attrib
.get('width')),
836 'height': int_or_none(r
.attrib
.get('height')),
837 'tbr': int_or_none(r
.attrib
.get('bandwidth'), 1000),
838 'asr': int_or_none(r
.attrib
.get('audioSamplingRate')),
839 'filesize': filesize
,
840 'fps': int_or_none(r
.attrib
.get('frameRate')),
843 existing_format
= next(
845 if fo
['format_id'] == format_id
)
846 except StopIteration:
847 full_info
= self
._formats
.get(format_id
, {}).copy()
849 codecs
= r
.attrib
.get('codecs')
851 if full_info
.get('acodec') == 'none' and 'vcodec' not in full_info
:
852 full_info
['vcodec'] = codecs
853 elif full_info
.get('vcodec') == 'none' and 'acodec' not in full_info
:
854 full_info
['acodec'] = codecs
855 formats
.append(full_info
)
857 existing_format
.update(f
)
859 self
.report_warning('Unknown MIME type %s in DASH manifest' % mime_type
)
862 def _real_extract(self
, url
):
864 'http' if self
._downloader
.params
.get('prefer_insecure', False)
867 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
868 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
870 url
= proto
+ '://www.youtube.com/' + compat_urllib_parse_unquote(mobj
.group(1)).lstrip('/')
871 video_id
= self
.extract_id(url
)
874 url
= proto
+ '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
875 video_webpage
= self
._download
_webpage
(url
, video_id
)
877 # Attempt to extract SWF player URL
878 mobj
= re
.search(r
'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
880 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
886 def add_dash_mpd(video_info
):
887 dash_mpd
= video_info
.get('dashmpd')
888 if dash_mpd
and dash_mpd
[0] not in dash_mpds
:
889 dash_mpds
.append(dash_mpd
[0])
893 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
895 # We simulate the access to the video from www.youtube.com/v/{video_id}
896 # this can be viewed without login into Youtube
897 url
= proto
+ '://www.youtube.com/embed/%s' % video_id
898 embed_webpage
= self
._download
_webpage
(url
, video_id
, 'Downloading embed webpage')
899 data
= compat_urllib_parse
.urlencode({
900 'video_id': video_id
,
901 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
902 'sts': self
._search
_regex
(
903 r
'"sts"\s*:\s*(\d+)', embed_webpage
, 'sts', default
=''),
905 video_info_url
= proto
+ '://www.youtube.com/get_video_info?' + data
906 video_info_webpage
= self
._download
_webpage
(
907 video_info_url
, video_id
,
908 note
='Refetching age-gated info webpage',
909 errnote
='unable to download video info webpage')
910 video_info
= compat_parse_qs(video_info_webpage
)
911 add_dash_mpd(video_info
)
915 # Try looking directly into the video webpage
916 mobj
= re
.search(r
';ytplayer\.config\s*=\s*({.*?});', video_webpage
)
918 json_code
= uppercase_escape(mobj
.group(1))
919 ytplayer_config
= json
.loads(json_code
)
920 args
= ytplayer_config
['args']
921 if args
.get('url_encoded_fmt_stream_map'):
922 # Convert to the same format returned by compat_parse_qs
923 video_info
= dict((k
, [v
]) for k
, v
in args
.items())
924 add_dash_mpd(video_info
)
925 if not video_info
or self
._downloader
.params
.get('youtube_include_dash_manifest', True):
926 # We also try looking in get_video_info since it may contain different dashmpd
927 # URL that points to a DASH manifest with possibly different itag set (some itags
928 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
929 # manifest pointed by get_video_info's dashmpd).
930 # The general idea is to take a union of itags of both DASH manifests (for example
931 # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
932 self
.report_video_info_webpage_download(video_id
)
933 for el_type
in ['&el=info', '&el=embedded', '&el=detailpage', '&el=vevo', '']:
935 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
936 % (proto
, video_id
, el_type
))
937 video_info_webpage
= self
._download
_webpage
(
939 video_id
, note
=False,
940 errnote
='unable to download video info webpage')
941 get_video_info
= compat_parse_qs(video_info_webpage
)
942 add_dash_mpd(get_video_info
)
944 video_info
= get_video_info
945 if 'token' in get_video_info
:
947 if 'token' not in video_info
:
948 if 'reason' in video_info
:
949 if 'The uploader has not made this video available in your country.' in video_info
['reason']:
950 regions_allowed
= self
._html
_search
_meta
('regionsAllowed', video_webpage
, default
=None)
951 if regions_allowed
is not None:
952 raise ExtractorError('YouTube said: This video is available in %s only' % (
953 ', '.join(map(ISO3166Utils
.short2full
, regions_allowed
.split(',')))),
955 raise ExtractorError(
956 'YouTube said: %s' % video_info
['reason'][0],
957 expected
=True, video_id
=video_id
)
959 raise ExtractorError(
960 '"token" parameter not in video info for unknown reason',
963 if 'view_count' in video_info
:
964 view_count
= int(video_info
['view_count'][0])
968 # Check for "rental" videos
969 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
970 raise ExtractorError('"rental" videos not supported')
972 # Start extracting information
973 self
.report_information_extraction(video_id
)
976 if 'author' not in video_info
:
977 raise ExtractorError('Unable to extract uploader name')
978 video_uploader
= compat_urllib_parse_unquote_plus(video_info
['author'][0])
981 video_uploader_id
= None
982 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
984 video_uploader_id
= mobj
.group(1)
986 self
._downloader
.report_warning('unable to extract uploader nickname')
989 if 'title' in video_info
:
990 video_title
= video_info
['title'][0]
992 self
._downloader
.report_warning('Unable to extract video title')
996 # We try first to get a high quality image:
997 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
998 video_webpage
, re
.DOTALL
)
999 if m_thumb
is not None:
1000 video_thumbnail
= m_thumb
.group(1)
1001 elif 'thumbnail_url' not in video_info
:
1002 self
._downloader
.report_warning('unable to extract video thumbnail')
1003 video_thumbnail
= None
1004 else: # don't panic if we can't find it
1005 video_thumbnail
= compat_urllib_parse_unquote_plus(video_info
['thumbnail_url'][0])
1008 upload_date
= self
._html
_search
_meta
(
1009 'datePublished', video_webpage
, 'upload date', default
=None)
1011 upload_date
= self
._search
_regex
(
1012 [r
'(?s)id="eow-date.*?>(.*?)</span>',
1013 r
'id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live|Started) on (.+?)</strong>'],
1014 video_webpage
, 'upload date', default
=None)
1016 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
1017 upload_date
= unified_strdate(upload_date
)
1019 m_cat_container
= self
._search
_regex
(
1020 r
'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
1021 video_webpage
, 'categories', default
=None)
1023 category
= self
._html
_search
_regex
(
1024 r
'(?s)<a[^<]+>(.*?)</a>', m_cat_container
, 'category',
1026 video_categories
= None if category
is None else [category
]
1028 video_categories
= None
1031 video_description
= get_element_by_id("eow-description", video_webpage
)
1032 if video_description
:
1033 video_description
= re
.sub(r
'''(?x)
1035 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1037 (?:[a-zA-Z-]+="[^"]+"\s+)*?
1038 class="yt-uix-redirect-link"\s*>
1041 ''', r
'\1', video_description
)
1042 video_description
= clean_html(video_description
)
1044 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
1046 video_description
= unescapeHTML(fd_mobj
.group(1))
1048 video_description
= ''
1050 def _extract_count(count_name
):
1051 return str_to_int(self
._search
_regex
(
1052 r
'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
1053 % re
.escape(count_name
),
1054 video_webpage
, count_name
, default
=None))
1056 like_count
= _extract_count('like')
1057 dislike_count
= _extract_count('dislike')
1060 video_subtitles
= self
.extract_subtitles(video_id
, video_webpage
)
1061 automatic_captions
= self
.extract_automatic_captions(video_id
, video_webpage
)
1063 if 'length_seconds' not in video_info
:
1064 self
._downloader
.report_warning('unable to extract video duration')
1065 video_duration
= None
1067 video_duration
= int(compat_urllib_parse_unquote_plus(video_info
['length_seconds'][0]))
1070 video_annotations
= None
1071 if self
._downloader
.params
.get('writeannotations', False):
1072 video_annotations
= self
._extract
_annotations
(video_id
)
1074 def _map_to_format_list(urlmap
):
1076 for itag
, video_real_url
in urlmap
.items():
1079 'url': video_real_url
,
1080 'player_url': player_url
,
1082 if itag
in self
._formats
:
1083 dct
.update(self
._formats
[itag
])
1087 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
1088 self
.report_rtmp_download()
1090 'format_id': '_rtmp',
1092 'url': video_info
['conn'][0],
1093 'player_url': player_url
,
1095 elif len(video_info
.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info
.get('adaptive_fmts', [''])[0]) >= 1:
1096 encoded_url_map
= video_info
.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info
.get('adaptive_fmts', [''])[0]
1097 if 'rtmpe%3Dyes' in encoded_url_map
:
1098 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
1100 for url_data_str
in encoded_url_map
.split(','):
1101 url_data
= compat_parse_qs(url_data_str
)
1102 if 'itag' not in url_data
or 'url' not in url_data
:
1104 format_id
= url_data
['itag'][0]
1105 url
= url_data
['url'][0]
1107 if 'sig' in url_data
:
1108 url
+= '&signature=' + url_data
['sig'][0]
1109 elif 's' in url_data
:
1110 encrypted_sig
= url_data
['s'][0]
1111 ASSETS_RE
= r
'"assets":.+?"js":\s*("[^"]+")'
1113 jsplayer_url_json
= self
._search
_regex
(
1115 embed_webpage
if age_gate
else video_webpage
,
1116 'JS player URL (1)', default
=None)
1117 if not jsplayer_url_json
and not age_gate
:
1118 # We need the embed website after all
1119 if embed_webpage
is None:
1120 embed_url
= proto
+ '://www.youtube.com/embed/%s' % video_id
1121 embed_webpage
= self
._download
_webpage
(
1122 embed_url
, video_id
, 'Downloading embed webpage')
1123 jsplayer_url_json
= self
._search
_regex
(
1124 ASSETS_RE
, embed_webpage
, 'JS player URL')
1126 player_url
= json
.loads(jsplayer_url_json
)
1127 if player_url
is None:
1128 player_url_json
= self
._search
_regex
(
1129 r
'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1130 video_webpage
, 'age gate player URL')
1131 player_url
= json
.loads(player_url_json
)
1133 if self
._downloader
.params
.get('verbose'):
1134 if player_url
is None:
1135 player_version
= 'unknown'
1136 player_desc
= 'unknown'
1138 if player_url
.endswith('swf'):
1139 player_version
= self
._search
_regex
(
1140 r
'-(.+?)(?:/watch_as3)?\.swf$', player_url
,
1141 'flash player', fatal
=False)
1142 player_desc
= 'flash player %s' % player_version
1144 player_version
= self
._search
_regex
(
1145 r
'html5player-([^/]+?)(?:/html5player)?\.js',
1147 'html5 player', fatal
=False)
1148 player_desc
= 'html5 player %s' % player_version
1150 parts_sizes
= self
._signature
_cache
_id
(encrypted_sig
)
1151 self
.to_screen('{%s} signature length %s, %s' %
1152 (format_id
, parts_sizes
, player_desc
))
1154 signature
= self
._decrypt
_signature
(
1155 encrypted_sig
, video_id
, player_url
, age_gate
)
1156 url
+= '&signature=' + signature
1157 if 'ratebypass' not in url
:
1158 url
+= '&ratebypass=yes'
1159 url_map
[format_id
] = url
1160 formats
= _map_to_format_list(url_map
)
1161 elif video_info
.get('hlsvp'):
1162 manifest_url
= video_info
['hlsvp'][0]
1163 url_map
= self
._extract
_from
_m
3u8(manifest_url
, video_id
)
1164 formats
= _map_to_format_list(url_map
)
1166 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1168 # Look for the DASH manifest
1169 if self
._downloader
.params
.get('youtube_include_dash_manifest', True):
1170 dash_mpd_fatal
= True
1171 for dash_manifest_url
in dash_mpds
:
1174 for df
in self
._parse
_dash
_manifest
(
1175 video_id
, dash_manifest_url
, player_url
, age_gate
, dash_mpd_fatal
):
1176 # Do not overwrite DASH format found in some previous DASH manifest
1177 if df
['format_id'] not in dash_formats
:
1178 dash_formats
[df
['format_id']] = df
1179 # Additional DASH manifests may end up in HTTP Error 403 therefore
1180 # allow them to fail without bug report message if we already have
1181 # some DASH manifest succeeded. This is temporary workaround to reduce
1182 # burst of bug reports until we figure out the reason and whether it
1183 # can be fixed at all.
1184 dash_mpd_fatal
= False
1185 except (ExtractorError
, KeyError) as e
:
1186 self
.report_warning(
1187 'Skipping DASH manifest: %r' % e
, video_id
)
1189 # Remove the formats we found through non-DASH, they
1190 # contain less info and it can be wrong, because we use
1191 # fixed values (for example the resolution). See
1192 # https://github.com/rg3/youtube-dl/issues/5774 for an
1194 formats
= [f
for f
in formats
if f
['format_id'] not in dash_formats
.keys()]
1195 formats
.extend(dash_formats
.values())
1197 # Check for malformed aspect ratio
1198 stretched_m
= re
.search(
1199 r
'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1202 ratio
= float(stretched_m
.group('w')) / float(stretched_m
.group('h'))
1204 if f
.get('vcodec') != 'none':
1205 f
['stretched_ratio'] = ratio
1207 self
._sort
_formats
(formats
)
1211 'uploader': video_uploader
,
1212 'uploader_id': video_uploader_id
,
1213 'upload_date': upload_date
,
1214 'title': video_title
,
1215 'thumbnail': video_thumbnail
,
1216 'description': video_description
,
1217 'categories': video_categories
,
1218 'subtitles': video_subtitles
,
1219 'automatic_captions': automatic_captions
,
1220 'duration': video_duration
,
1221 'age_limit': 18 if age_gate
else 0,
1222 'annotations': video_annotations
,
1223 'webpage_url': proto
+ '://www.youtube.com/watch?v=%s' % video_id
,
1224 'view_count': view_count
,
1225 'like_count': like_count
,
1226 'dislike_count': dislike_count
,
1227 'average_rating': float_or_none(video_info
.get('avg_rating', [None])[0]),
1232 class YoutubePlaylistIE(YoutubeBaseInfoExtractor
):
1233 IE_DESC
= 'YouTube.com playlists'
1234 _VALID_URL
= r
"""(?x)(?:
1239 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1240 \? (?:.*?&)*? (?:p|a|list)=
1244 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1245 # Top tracks, they can also include dots
1250 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1252 _TEMPLATE_URL
= 'https://www.youtube.com/playlist?list=%s'
1253 _VIDEO_RE
= r
'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1254 IE_NAME
= 'youtube:playlist'
1256 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1258 'title': 'ytdl test PL',
1259 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1261 'playlist_count': 3,
1263 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1265 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1266 'title': 'YDL_Empty_List',
1268 'playlist_count': 0,
1270 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1271 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1273 'title': '29C3: Not my department',
1274 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1276 'playlist_count': 95,
1278 'note': 'issue #673',
1279 'url': 'PLBB231211A4F62143',
1281 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1282 'id': 'PLBB231211A4F62143',
1284 'playlist_mincount': 26,
1286 'note': 'Large playlist',
1287 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1289 'title': 'Uploads from Cauchemar',
1290 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1292 'playlist_mincount': 799,
1294 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1296 'title': 'YDL_safe_search',
1297 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1299 'playlist_count': 2,
1302 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1303 'playlist_count': 4,
1306 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1309 'note': 'Embedded SWF player',
1310 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1311 'playlist_count': 4,
1314 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1317 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1318 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1320 'title': 'Uploads from Interstellar Movie',
1321 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1323 'playlist_mincout': 21,
1326 def _real_initialize(self
):
1329 def _extract_mix(self
, playlist_id
):
1330 # The mixes are generated from a single video
1331 # the id of the playlist is just 'RD' + video_id
1332 url
= 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id
[-11:], playlist_id
)
1333 webpage
= self
._download
_webpage
(
1334 url
, playlist_id
, 'Downloading Youtube mix')
1335 search_title
= lambda class_name
: get_element_by_attribute('class', class_name
, webpage
)
1337 search_title('playlist-title') or
1338 search_title('title long-title') or
1339 search_title('title'))
1340 title
= clean_html(title_span
)
1341 ids
= orderedSet(re
.findall(
1342 r
'''(?xs)data-video-username=".*?".*?
1343 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re
.escape(playlist_id
),
1345 url_results
= self
._ids
_to
_results
(ids
)
1347 return self
.playlist_result(url_results
, playlist_id
, title
)
1349 def _extract_playlist(self
, playlist_id
):
1350 url
= self
._TEMPLATE
_URL
% playlist_id
1351 page
= self
._download
_webpage
(url
, playlist_id
)
1353 for match
in re
.findall(r
'<div class="yt-alert-message">([^<]+)</div>', page
):
1354 match
= match
.strip()
1355 # Check if the playlist exists or is private
1356 if re
.match(r
'[^<]*(The|This) playlist (does not exist|is private)[^<]*', match
):
1357 raise ExtractorError(
1358 'The playlist doesn\'t exist or is private, use --username or '
1359 '--netrc to access it.',
1361 elif re
.match(r
'[^<]*Invalid parameters[^<]*', match
):
1362 raise ExtractorError(
1363 'Invalid parameters. Maybe URL is incorrect.',
1365 elif re
.match(r
'[^<]*Choose your language[^<]*', match
):
1368 self
.report_warning('Youtube gives an alert message: ' + match
)
1370 # Extract the video ids from the playlist pages
1372 more_widget_html
= content_html
= page
1373 for page_num
in itertools
.count(1):
1374 matches
= re
.finditer(self
._VIDEO
_RE
, content_html
)
1375 # We remove the duplicates and the link with index 0
1376 # (it's not the first video of the playlist)
1377 new_ids
= orderedSet(m
.group('id') for m
in matches
if m
.group('index') != '0')
1378 for vid_id
in new_ids
:
1379 yield self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
1381 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1385 more
= self
._download
_json
(
1386 'https://youtube.com/%s' % mobj
.group('more'), playlist_id
,
1387 'Downloading page #%s' % page_num
,
1388 transform_source
=uppercase_escape
)
1389 content_html
= more
['content_html']
1390 if not content_html
.strip():
1391 # Some webpages show a "Load more" button but they don't
1394 more_widget_html
= more
['load_more_widget_html']
1396 playlist_title
= self
._html
_search
_regex
(
1397 r
'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1400 return self
.playlist_result(_entries(), playlist_id
, playlist_title
)
1402 def _real_extract(self
, url
):
1403 # Extract playlist id
1404 mobj
= re
.match(self
._VALID
_URL
, url
)
1406 raise ExtractorError('Invalid URL: %s' % url
)
1407 playlist_id
= mobj
.group(1) or mobj
.group(2)
1409 # Check if it's a video-specific URL
1410 query_dict
= compat_urlparse
.parse_qs(compat_urlparse
.urlparse(url
).query
)
1411 if 'v' in query_dict
:
1412 video_id
= query_dict
['v'][0]
1413 if self
._downloader
.params
.get('noplaylist'):
1414 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1415 return self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1417 self
.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id
, video_id
))
1419 if playlist_id
.startswith('RD') or playlist_id
.startswith('UL'):
1420 # Mixes require a custom extraction process
1421 return self
._extract
_mix
(playlist_id
)
1423 return self
._extract
_playlist
(playlist_id
)
1426 class YoutubeChannelIE(InfoExtractor
):
1427 IE_DESC
= 'YouTube.com channels'
1428 _VALID_URL
= r
'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1429 _TEMPLATE_URL
= 'https://www.youtube.com/channel/%s/videos'
1430 IE_NAME
= 'youtube:channel'
1432 'note': 'paginated channel',
1433 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1434 'playlist_mincount': 91,
1436 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1441 def extract_videos_from_page(page
):
1444 for mobj
in re
.finditer(r
'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page
):
1445 video_id
= mobj
.group('id')
1446 video_title
= unescapeHTML(mobj
.group('title'))
1448 idx
= ids_in_page
.index(video_id
)
1449 if video_title
and not titles_in_page
[idx
]:
1450 titles_in_page
[idx
] = video_title
1452 ids_in_page
.append(video_id
)
1453 titles_in_page
.append(video_title
)
1454 return zip(ids_in_page
, titles_in_page
)
1456 def _real_extract(self
, url
):
1457 channel_id
= self
._match
_id
(url
)
1459 url
= self
._TEMPLATE
_URL
% channel_id
1461 # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
1462 # Workaround by extracting as a playlist if managed to obtain channel playlist URL
1463 # otherwise fallback on channel by page extraction
1464 channel_page
= self
._download
_webpage
(
1465 url
+ '?view=57', channel_id
,
1466 'Downloading channel page', fatal
=False)
1467 channel_playlist_id
= self
._html
_search
_meta
(
1468 'channelId', channel_page
, 'channel id', default
=None)
1469 if not channel_playlist_id
:
1470 channel_playlist_id
= self
._search
_regex
(
1471 r
'data-channel-external-id="([^"]+)"',
1472 channel_page
, 'channel id', default
=None)
1473 if channel_playlist_id
and channel_playlist_id
.startswith('UC'):
1474 playlist_id
= 'UU' + channel_playlist_id
[2:]
1475 return self
.url_result(
1476 compat_urlparse
.urljoin(url
, '/playlist?list=%s' % playlist_id
), 'YoutubePlaylist')
1478 channel_page
= self
._download
_webpage
(url
, channel_id
, 'Downloading page #1')
1479 autogenerated
= re
.search(r
'''(?x)
1481 channel-header-autogenerated-label|
1482 yt-channel-title-autogenerated
1483 )[^"]*"''', channel_page
) is not None
1486 # The videos are contained in a single page
1487 # the ajax pages can't be used, they are empty
1490 video_id
, 'Youtube', video_id
=video_id
,
1491 video_title
=video_title
)
1492 for video_id
, video_title
in self
.extract_videos_from_page(channel_page
)]
1493 return self
.playlist_result(entries
, channel_id
)
1496 more_widget_html
= content_html
= channel_page
1497 for pagenum
in itertools
.count(1):
1499 for video_id
, video_title
in self
.extract_videos_from_page(content_html
):
1500 yield self
.url_result(
1501 video_id
, 'Youtube', video_id
=video_id
,
1502 video_title
=video_title
)
1505 r
'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1510 more
= self
._download
_json
(
1511 'https://youtube.com/%s' % mobj
.group('more'), channel_id
,
1512 'Downloading page #%s' % (pagenum
+ 1),
1513 transform_source
=uppercase_escape
)
1514 content_html
= more
['content_html']
1515 more_widget_html
= more
['load_more_widget_html']
1517 return self
.playlist_result(_entries(), channel_id
)
1520 class YoutubeUserIE(YoutubeChannelIE
):
1521 IE_DESC
= 'YouTube.com user videos (URL or "ytuser" keyword)'
1522 _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_-]+)'
1523 _TEMPLATE_URL
= 'https://www.youtube.com/user/%s/videos'
1524 IE_NAME
= 'youtube:user'
1527 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1528 'playlist_mincount': 320,
1530 'title': 'TheLinuxFoundation',
1533 'url': 'ytuser:phihag',
1534 'only_matching': True,
1538 def suitable(cls
, url
):
1539 # Don't return True if the url can be extracted with other youtube
1540 # extractor, the regex would is too permissive and it would match.
1541 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1542 if any(ie
.suitable(url
) for ie
in other_ies
):
1545 return super(YoutubeUserIE
, cls
).suitable(url
)
1548 class YoutubeSearchIE(SearchInfoExtractor
, YoutubePlaylistIE
):
1549 IE_DESC
= 'YouTube.com searches'
1550 # there doesn't appear to be a real limit, for example if you search for
1551 # 'python' you get more than 8.000.000 results
1552 _MAX_RESULTS
= float('inf')
1553 IE_NAME
= 'youtube:search'
1554 _SEARCH_KEY
= 'ytsearch'
1555 _EXTRA_QUERY_ARGS
= {}
1558 def _get_n_results(self
, query
, n
):
1559 """Get a specified number of results for a query"""
1564 for pagenum
in itertools
.count(1):
1566 'search_query': query
.encode('utf-8'),
1570 url_query
.update(self
._EXTRA
_QUERY
_ARGS
)
1571 result_url
= 'https://www.youtube.com/results?' + compat_urllib_parse
.urlencode(url_query
)
1572 data
= self
._download
_json
(
1573 result_url
, video_id
='query "%s"' % query
,
1574 note
='Downloading page %s' % pagenum
,
1575 errnote
='Unable to download API page')
1576 html_content
= data
[1]['body']['content']
1578 if 'class="search-message' in html_content
:
1579 raise ExtractorError(
1580 '[youtube] No video results', expected
=True)
1582 new_videos
= self
._ids
_to
_results
(orderedSet(re
.findall(
1583 r
'href="/watch\?v=(.{11})', html_content
)))
1584 videos
+= new_videos
1585 if not new_videos
or len(videos
) > limit
:
1590 return self
.playlist_result(videos
, query
)
1593 class YoutubeSearchDateIE(YoutubeSearchIE
):
1594 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1595 _SEARCH_KEY
= 'ytsearchdate'
1596 IE_DESC
= 'YouTube.com searches, newest videos first'
1597 _EXTRA_QUERY_ARGS
= {'search_sort': 'video_date_uploaded'}
1600 class YoutubeSearchURLIE(InfoExtractor
):
1601 IE_DESC
= 'YouTube.com search URLs'
1602 IE_NAME
= 'youtube:search_url'
1603 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1605 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1606 'playlist_mincount': 5,
1608 'title': 'youtube-dl test video',
1612 def _real_extract(self
, url
):
1613 mobj
= re
.match(self
._VALID
_URL
, url
)
1614 query
= compat_urllib_parse_unquote_plus(mobj
.group('query'))
1616 webpage
= self
._download
_webpage
(url
, query
)
1617 result_code
= self
._search
_regex
(
1618 r
'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage
, 'result HTML')
1620 part_codes
= re
.findall(
1621 r
'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code
)
1623 for part_code
in part_codes
:
1624 part_title
= self
._html
_search
_regex
(
1625 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1626 part_url_snippet
= self
._html
_search
_regex
(
1627 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1628 part_url
= compat_urlparse
.urljoin(
1629 'https://www.youtube.com/', part_url_snippet
)
1633 'title': part_title
,
1637 '_type': 'playlist',
1643 class YoutubeShowIE(InfoExtractor
):
1644 IE_DESC
= 'YouTube.com (multi-season) shows'
1645 _VALID_URL
= r
'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1646 IE_NAME
= 'youtube:show'
1648 'url': 'http://www.youtube.com/show/airdisasters',
1649 'playlist_mincount': 3,
1651 'id': 'airdisasters',
1652 'title': 'Air Disasters',
1656 def _real_extract(self
, url
):
1657 mobj
= re
.match(self
._VALID
_URL
, url
)
1658 playlist_id
= mobj
.group('id')
1659 webpage
= self
._download
_webpage
(
1660 url
, playlist_id
, 'Downloading show webpage')
1661 # There's one playlist for each season of the show
1662 m_seasons
= list(re
.finditer(r
'href="(/playlist\?list=.*?)"', webpage
))
1663 self
.to_screen('%s: Found %s seasons' % (playlist_id
, len(m_seasons
)))
1666 'https://www.youtube.com' + season
.group(1), 'YoutubePlaylist')
1667 for season
in m_seasons
1669 title
= self
._og
_search
_title
(webpage
, fatal
=False)
1672 '_type': 'playlist',
1679 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1681 Base class for feed extractors
1682 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1684 _LOGIN_REQUIRED
= True
1688 return 'youtube:%s' % self
._FEED
_NAME
1690 def _real_initialize(self
):
1693 def _real_extract(self
, url
):
1694 page
= self
._download
_webpage
(
1695 'https://www.youtube.com/feed/%s' % self
._FEED
_NAME
, self
._PLAYLIST
_TITLE
)
1697 # The extraction process is the same as for playlists, but the regex
1698 # for the video ids doesn't contain an index
1700 more_widget_html
= content_html
= page
1701 for page_num
in itertools
.count(1):
1702 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
1704 # 'recommended' feed has infinite 'load more' and each new portion spins
1705 # the same videos in (sometimes) slightly different order, so we'll check
1706 # for unicity and break when portion has no new videos
1707 new_ids
= filter(lambda video_id
: video_id
not in ids
, orderedSet(matches
))
1713 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1717 more
= self
._download
_json
(
1718 'https://youtube.com/%s' % mobj
.group('more'), self
._PLAYLIST
_TITLE
,
1719 'Downloading page #%s' % page_num
,
1720 transform_source
=uppercase_escape
)
1721 content_html
= more
['content_html']
1722 more_widget_html
= more
['load_more_widget_html']
1724 return self
.playlist_result(
1725 self
._ids
_to
_results
(ids
), playlist_title
=self
._PLAYLIST
_TITLE
)
1728 class YoutubeWatchLaterIE(YoutubePlaylistIE
):
1729 IE_NAME
= 'youtube:watchlater'
1730 IE_DESC
= 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1731 _VALID_URL
= r
'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1733 _TESTS
= [] # override PlaylistIE tests
1735 def _real_extract(self
, url
):
1736 return self
._extract
_playlist
('WL')
1739 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
1740 IE_NAME
= 'youtube:favorites'
1741 IE_DESC
= 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1742 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1743 _LOGIN_REQUIRED
= True
1745 def _real_extract(self
, url
):
1746 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1747 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, 'favourites playlist id')
1748 return self
.url_result(playlist_id
, 'YoutubePlaylist')
1751 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
1752 IE_DESC
= 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1753 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1754 _FEED_NAME
= 'recommended'
1755 _PLAYLIST_TITLE
= 'Youtube Recommended videos'
1758 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor
):
1759 IE_DESC
= 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1760 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1761 _FEED_NAME
= 'subscriptions'
1762 _PLAYLIST_TITLE
= 'Youtube Subscriptions'
1765 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
1766 IE_DESC
= 'Youtube watch history, ":ythistory" for short (requires authentication)'
1767 _VALID_URL
= 'https?://www\.youtube\.com/feed/history|:ythistory'
1768 _FEED_NAME
= 'history'
1769 _PLAYLIST_TITLE
= 'Youtube History'
1772 class YoutubeTruncatedURLIE(InfoExtractor
):
1773 IE_NAME
= 'youtube:truncated_url'
1774 IE_DESC
= False # Do not list
1775 _VALID_URL
= r
'''(?x)
1777 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1780 annotation_id=annotation_[^&]+|
1785 attribution_link\?a=[^&]+
1791 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1792 'only_matching': True,
1794 'url': 'http://www.youtube.com/watch?',
1795 'only_matching': True,
1797 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1798 'only_matching': True,
1800 'url': 'https://www.youtube.com/watch?feature=foo',
1801 'only_matching': True,
1803 'url': 'https://www.youtube.com/watch?hl=en-GB',
1804 'only_matching': True,
1807 def _real_extract(self
, url
):
1808 raise ExtractorError(
1809 'Did you forget to quote the URL? Remember that & is a meta '
1810 'character in most shells, so you want to put the URL in quotes, '
1812 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1813 ' or simply youtube-dl BaW_jenozKc .',
1817 class YoutubeTruncatedIDIE(InfoExtractor
):
1818 IE_NAME
= 'youtube:truncated_id'
1819 IE_DESC
= False # Do not list
1820 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1823 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1824 'only_matching': True,
1827 def _real_extract(self
, url
):
1828 video_id
= self
._match
_id
(url
)
1829 raise ExtractorError(
1830 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id
, url
),