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_request
,
28 get_element_by_attribute
,
38 class YoutubeBaseInfoExtractor(InfoExtractor
):
39 """Provide base functions for Youtube extractors"""
40 _LOGIN_URL
= 'https://accounts.google.com/ServiceLogin'
41 _TWOFACTOR_URL
= 'https://accounts.google.com/SecondFactor'
42 _NETRC_MACHINE
= 'youtube'
43 # If True it will raise an error if no login info is provided
44 _LOGIN_REQUIRED
= False
46 def _set_language(self
):
48 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
49 # YouTube sets the expire time to about two months
50 expire_time
=time
.time() + 2 * 30 * 24 * 3600)
54 Attempt to log in to YouTube.
55 True is returned if successful or skipped.
56 False is returned if login failed.
58 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
60 (username
, password
) = self
._get
_login
_info
()
61 # No authentication to be performed
63 if self
._LOGIN
_REQUIRED
:
64 raise ExtractorError('No login info available, needed for using %s.' % self
.IE_NAME
, expected
=True)
67 login_page
= self
._download
_webpage
(
68 self
._LOGIN
_URL
, None,
69 note
='Downloading login page',
70 errnote
='unable to fetch login page', fatal
=False)
71 if login_page
is False:
74 galx
= self
._search
_regex
(r
'(?s)<input.+?name="GALX".+?value="(.+?)"',
75 login_page
, 'Login GALX parameter')
79 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
84 'PersistentCookie': 'yes',
86 'bgresponse': 'js_disabled',
87 'checkConnection': '',
88 'checkedDomains': 'youtube',
100 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
102 login_form
= dict((k
.encode('utf-8'), v
.encode('utf-8')) for k
, v
in login_form_strs
.items())
103 login_data
= compat_urllib_parse
.urlencode(login_form
).encode('ascii')
105 req
= compat_urllib_request
.Request(self
._LOGIN
_URL
, login_data
)
106 login_results
= self
._download
_webpage
(
108 note
='Logging in', errnote
='unable to log in', fatal
=False)
109 if login_results
is False:
112 if re
.search(r
'id="errormsg_0_Passwd"', login_results
) is not None:
113 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected
=True)
116 # TODO add SMS and phone call support - these require making a request and then prompting the user
118 if re
.search(r
'(?i)<form[^>]* id="gaia_secondfactorform"', login_results
) is not None:
119 tfa_code
= self
._get
_tfa
_info
()
122 self
._downloader
.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
123 self
._downloader
.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
126 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
128 match
= re
.search(r
'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
130 self._downloader.report_warning('Failed to get secTok
- did the page structure change?
')
131 secTok = match.group(1)
132 match = re.search(r'id="timeStmp"\n\s
+value
=\'(.+)\'/>', login_results, re.M | re.U)
134 self._downloader.report_warning('Failed to get timeStmp
- did the page structure change?
')
135 timeStmp = match.group(1)
138 'continue': 'https
://www
.youtube
.com
/signin?action_handle_signin
=true
&feature
=sign_in_button
&hl
=en_US
&nomobiletemp
=1',
140 'smsUserPin
': tfa_code,
141 'smsVerifyPin
': 'Verify
',
143 'PersistentCookie
': 'yes
',
144 'checkConnection
': '',
145 'checkedDomains
': 'youtube
',
148 'timeStmp
': timeStmp,
149 'service
': 'youtube
',
152 tfa_form = dict((k.encode('utf
-8'), v.encode('utf
-8')) for k, v in tfa_form_strs.items())
153 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii
')
155 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
156 tfa_results = self._download_webpage(
158 note='Submitting TFA code
', errnote='unable to submit tfa
', fatal=False)
160 if tfa_results is False:
163 if re.search(r'(?i
)<form
[^
>]* id="gaia_secondfactorform"', tfa_results) is not None:
164 self._downloader.report_warning('Two
-factor code expired
. Please
try again
, or use a one
-use backup code instead
.')
166 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', tfa_results) is not None:
167 self._downloader.report_warning('unable to log
in - did the page structure change?
')
169 if re.search(r'smsauth
-interstitial
-reviewsettings
', tfa_results) is not None:
170 self._downloader.report_warning('Your Google account has a security notice
. Please log
in on your web browser
, resolve the notice
, and try again
.')
173 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', login_results) is not None:
174 self._downloader.report_warning('unable to log
in: bad username
or password
')
178 def _real_initialize(self):
179 if self._downloader is None:
182 if not self._login():
186 class YoutubeIE(YoutubeBaseInfoExtractor):
187 IE_DESC = 'YouTube
.com
'
188 _VALID_URL = r"""(?x)^
190 (?:https?://|//) # http(s):// or protocol-independent URL
191 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
192 (?:www\.)?deturl\.com/www\.youtube\.com/|
193 (?:www\.)?pwnyoutube\.com/|
194 (?:www\.)?yourepeat\.com/|
195 tube\.majestyc\.net/|
196 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
197 (?:.*?\#/)? # handle anchor (#/) redirect urls
198 (?: # the various things that can precede the ID:
199 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
200 |(?: # or the v= param in all its forms
201 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
202 (?:\?|\#!?) # the params delimiter ? or # or #!
203 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
207 |youtu\.be/ # just youtu.be/xxxx
208 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
210 )? # all until now is optional -> you can pass the naked ID
211 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
212 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
213 (?(1).+)? # if we found the ID, everything can follow
215 _NEXT_URL_RE = r'[\?&]next_url
=([^
&]+)'
217 '5': {'ext': 'flv', 'width': 400, 'height': 240},
218 '6': {'ext': 'flv', 'width': 450, 'height': 270},
219 '13': {'ext': '3gp'},
220 '17': {'ext': '3gp', 'width': 176, 'height': 144},
221 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
222 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
223 '34': {'ext': 'flv', 'width': 640, 'height': 360},
224 '35': {'ext': 'flv', 'width': 854, 'height': 480},
225 '36': {'ext': '3gp', 'width': 320, 'height': 240},
226 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
227 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
228 '43': {'ext': 'webm', 'width': 640, 'height': 360},
229 '44': {'ext': 'webm', 'width': 854, 'height': 480},
230 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
231 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
235 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
236 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
237 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
238 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
239 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
240 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
241 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
243 # Apple HTTP Live Streaming
244 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
245 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
246 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
247 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
248 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
249 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
250 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
253 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
254 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
255 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
256 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
257 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
258 '138': {'ext': 'mp4', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
259 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
260 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
261 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
262 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
263 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
266 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50, 'container': 'm4a_dash'},
267 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50, 'container': 'm4a_dash'},
268 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 256, 'preference': -50, 'container': 'm4a_dash'},
271 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
272 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
273 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
274 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
275 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
276 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
277 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
278 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
279 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
280 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
281 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
282 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
283 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
284 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
285 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
286 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
287 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
288 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
289 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
290 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
291 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
294 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
295 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
297 # Dash webm audio with opus inside
298 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
299 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
300 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
303 '_rtmp
': {'protocol': 'rtmp'},
309 'url
': 'http
://www
.youtube
.com
/watch?v
=BaW_jenozKc
',
313 'title
': 'youtube
-dl test video
"\'/\\ä↭𝕐',
314 'uploader': 'Philipp Hagemeister',
315 'uploader_id': 'phihag',
316 'upload_date': '20121002',
317 '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 .',
318 'categories
': ['Science
& Technology
'],
320 'dislike_count
': int,
324 'url
': 'http
://www
.youtube
.com
/watch?v
=UxxajLWwzqY
',
325 'note
': 'Test generic use_cipher_signature
video (#897)',
329 'upload_date': '20120506',
330 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
331 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
332 'uploader': 'Icona Pop',
333 'uploader_id': 'IconaPop',
337 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
338 'note': 'Test VEVO video with age protection (#956)',
342 'upload_date': '20130703',
343 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
344 'description': 'md5:64249768eec3bc4276236606ea996373',
345 'uploader': 'justintimberlakeVEVO',
346 'uploader_id': 'justintimberlakeVEVO',
350 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
351 'note': 'Embed-only video (#1746)',
355 'upload_date': '20120608',
356 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
357 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
358 'uploader': 'SET India',
359 'uploader_id': 'setindia'
363 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
364 'note': '256k DASH audio (format 141) via DASH manifest',
368 'upload_date': '20121002',
369 'uploader_id': '8KVIDEO',
371 'uploader': '8KVIDEO',
372 'title': 'UHDTV TEST 8K VIDEO.mp4'
375 'youtube_include_dash_manifest': True,
379 # DASH manifest with encrypted signature
381 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
385 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
386 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
387 'uploader': 'AfrojackVEVO',
388 'uploader_id': 'AfrojackVEVO',
389 'upload_date': '20131011',
392 'youtube_include_dash_manifest': True,
396 # JS player signature function name containing $
398 'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
402 'title': 'Taylor Swift - Shake It Off',
403 'description': 'md5:2acfda1b285bdd478ccec22f9918199d',
404 'uploader': 'TaylorSwiftVEVO',
405 'uploader_id': 'TaylorSwiftVEVO',
406 'upload_date': '20140818',
409 'youtube_include_dash_manifest': True,
415 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
419 'upload_date': '20100909',
420 'uploader': 'The Amazing Atheist',
421 'uploader_id': 'TheAmazingAtheist',
422 'title': 'Burning Everyone\'s Koran',
423 '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',
426 # Normal age-gate video (No vevo, embed allowed)
428 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
432 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
433 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
434 'uploader': 'The Witcher',
435 'uploader_id': 'WitcherGame',
436 'upload_date': '20140605',
439 # Age-gate video with encrypted signature
441 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
445 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
446 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
447 'uploader': 'LloydVEVO',
448 'uploader_id': 'LloydVEVO',
449 'upload_date': '20110629',
452 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
454 'url': '__2ABJjxzNo',
458 'upload_date': '20100430',
459 'uploader_id': 'deadmau5',
460 'description': 'md5:12c56784b8032162bb936a5f76d55360',
461 'uploader': 'deadmau5',
462 'title': 'Deadmau5 - Some Chords (HD)',
464 'expected_warnings': [
465 'DASH manifest missing',
468 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
470 'url': 'lqQg6PlCWgI',
474 'upload_date': '20120731',
475 'uploader_id': 'olympic',
476 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
477 'uploader': 'Olympics',
478 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
481 'skip_download': 'requires avconv',
486 'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
490 'stretched_ratio': 16 / 9.,
491 'upload_date': '20110310',
492 'uploader_id': 'AllenMeow',
493 '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',
495 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
498 # url_encoded_fmt_stream_map is empty string
500 'url': 'qEJwOuvDf7I',
504 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
506 'upload_date': '20150404',
507 'uploader_id': 'spbelect',
508 'uploader': 'Наблюдатели Петербурга',
511 'skip_download': 'requires avconv',
516 def __init__(self
, *args
, **kwargs
):
517 super(YoutubeIE
, self
).__init
__(*args
, **kwargs
)
518 self
._player
_cache
= {}
520 def report_video_info_webpage_download(self
, video_id
):
521 """Report attempt to download video info webpage."""
522 self
.to_screen('%s: Downloading video info webpage' % video_id
)
524 def report_information_extraction(self
, video_id
):
525 """Report attempt to extract video information."""
526 self
.to_screen('%s: Extracting video information' % video_id
)
528 def report_unavailable_format(self
, video_id
, format
):
529 """Report extracted video URL."""
530 self
.to_screen('%s: Format %s not available' % (video_id
, format
))
532 def report_rtmp_download(self
):
533 """Indicate the download will use the RTMP protocol."""
534 self
.to_screen('RTMP download detected')
536 def _signature_cache_id(self
, example_sig
):
537 """ Return a string representation of a signature """
538 return '.'.join(compat_str(len(part
)) for part
in example_sig
.split('.'))
540 def _extract_signature_function(self
, video_id
, player_url
, example_sig
):
542 r
'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
545 raise ExtractorError('Cannot identify player %r' % player_url
)
546 player_type
= id_m
.group('ext')
547 player_id
= id_m
.group('id')
549 # Read from filesystem cache
550 func_id
= '%s_%s_%s' % (
551 player_type
, player_id
, self
._signature
_cache
_id
(example_sig
))
552 assert os
.path
.basename(func_id
) == func_id
554 cache_spec
= self
._downloader
.cache
.load('youtube-sigfuncs', func_id
)
555 if cache_spec
is not None:
556 return lambda s
: ''.join(s
[i
] for i
in cache_spec
)
559 'Downloading player %s' % player_url
560 if self
._downloader
.params
.get('verbose') else
561 'Downloading %s player %s' % (player_type
, player_id
)
563 if player_type
== 'js':
564 code
= self
._download
_webpage
(
565 player_url
, video_id
,
567 errnote
='Download of %s failed' % player_url
)
568 res
= self
._parse
_sig
_js
(code
)
569 elif player_type
== 'swf':
570 urlh
= self
._request
_webpage
(
571 player_url
, video_id
,
573 errnote
='Download of %s failed' % player_url
)
575 res
= self
._parse
_sig
_swf
(code
)
577 assert False, 'Invalid player type %r' % player_type
579 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
580 cache_res
= res(test_string
)
581 cache_spec
= [ord(c
) for c
in cache_res
]
583 self
._downloader
.cache
.store('youtube-sigfuncs', func_id
, cache_spec
)
586 def _print_sig_code(self
, func
, example_sig
):
587 def gen_sig_code(idxs
):
588 def _genslice(start
, end
, step
):
589 starts
= '' if start
== 0 else str(start
)
590 ends
= (':%d' % (end
+ step
)) if end
+ step
>= 0 else ':'
591 steps
= '' if step
== 1 else (':%d' % step
)
592 return 's[%s%s%s]' % (starts
, ends
, steps
)
595 # Quelch pyflakes warnings - start will be set when step is set
596 start
= '(Never used)'
597 for i
, prev
in zip(idxs
[1:], idxs
[:-1]):
601 yield _genslice(start
, prev
, step
)
604 if i
- prev
in [-1, 1]:
613 yield _genslice(start
, i
, step
)
615 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
616 cache_res
= func(test_string
)
617 cache_spec
= [ord(c
) for c
in cache_res
]
618 expr_code
= ' + '.join(gen_sig_code(cache_spec
))
619 signature_id_tuple
= '(%s)' % (
620 ', '.join(compat_str(len(p
)) for p
in example_sig
.split('.')))
621 code
= ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
622 ' return %s\n') % (signature_id_tuple
, expr_code
)
623 self
.to_screen('Extracted signature function:\n' + code
)
625 def _parse_sig_js(self
, jscode
):
626 funcname
= self
._search
_regex
(
627 r
'\.sig\|\|([a-zA-Z0-9$]+)\(', jscode
,
628 'Initial JS player signature function name')
630 jsi
= JSInterpreter(jscode
)
631 initial_function
= jsi
.extract_function(funcname
)
632 return lambda s
: initial_function([s
])
634 def _parse_sig_swf(self
, file_contents
):
635 swfi
= SWFInterpreter(file_contents
)
636 TARGET_CLASSNAME
= 'SignatureDecipher'
637 searched_class
= swfi
.extract_class(TARGET_CLASSNAME
)
638 initial_function
= swfi
.extract_function(searched_class
, 'decipher')
639 return lambda s
: initial_function([s
])
641 def _decrypt_signature(self
, s
, video_id
, player_url
, age_gate
=False):
642 """Turn the encrypted s field into a working signature"""
644 if player_url
is None:
645 raise ExtractorError('Cannot decrypt signature without player_url')
647 if player_url
.startswith('//'):
648 player_url
= 'https:' + player_url
650 player_id
= (player_url
, self
._signature
_cache
_id
(s
))
651 if player_id
not in self
._player
_cache
:
652 func
= self
._extract
_signature
_function
(
653 video_id
, player_url
, s
655 self
._player
_cache
[player_id
] = func
656 func
= self
._player
_cache
[player_id
]
657 if self
._downloader
.params
.get('youtube_print_sig_code'):
658 self
._print
_sig
_code
(func
, s
)
660 except Exception as e
:
661 tb
= traceback
.format_exc()
662 raise ExtractorError(
663 'Signature extraction failed: ' + tb
, cause
=e
)
665 def _get_subtitles(self
, video_id
, webpage
):
667 subs_doc
= self
._download
_xml
(
668 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
,
669 video_id
, note
=False)
670 except ExtractorError
as err
:
671 self
._downloader
.report_warning('unable to download video subtitles: %s' % compat_str(err
))
675 for track
in subs_doc
.findall('track'):
676 lang
= track
.attrib
['lang_code']
677 if lang
in sub_lang_list
:
680 for ext
in ['sbv', 'vtt', 'srt']:
681 params
= compat_urllib_parse
.urlencode({
685 'name': track
.attrib
['name'].encode('utf-8'),
688 'url': 'https://www.youtube.com/api/timedtext?' + params
,
691 sub_lang_list
[lang
] = sub_formats
692 if not sub_lang_list
:
693 self
._downloader
.report_warning('video doesn\'t have subtitles')
697 def _get_automatic_captions(self
, video_id
, webpage
):
698 """We need the webpage for getting the captions url, pass it as an
699 argument to speed up the process."""
700 self
.to_screen('%s: Looking for automatic captions' % video_id
)
701 mobj
= re
.search(r
';ytplayer.config = ({.*?});', webpage
)
702 err_msg
= 'Couldn\'t find automatic captions for %s' % video_id
704 self
._downloader
.report_warning(err_msg
)
706 player_config
= json
.loads(mobj
.group(1))
708 args
= player_config
['args']
709 caption_url
= args
['ttsurl']
710 timestamp
= args
['timestamp']
711 # We get the available subtitles
712 list_params
= compat_urllib_parse
.urlencode({
717 list_url
= caption_url
+ '&' + list_params
718 caption_list
= self
._download
_xml
(list_url
, video_id
)
719 original_lang_node
= caption_list
.find('track')
720 if original_lang_node
is None:
721 self
._downloader
.report_warning('Video doesn\'t have automatic captions')
723 original_lang
= original_lang_node
.attrib
['lang_code']
724 caption_kind
= original_lang_node
.attrib
.get('kind', '')
727 for lang_node
in caption_list
.findall('target'):
728 sub_lang
= lang_node
.attrib
['lang_code']
730 for ext
in ['sbv', 'vtt', 'srt']:
731 params
= compat_urllib_parse
.urlencode({
732 'lang': original_lang
,
736 'kind': caption_kind
,
739 'url': caption_url
+ '&' + params
,
742 sub_lang_list
[sub_lang
] = sub_formats
744 # An extractor error can be raise by the download process if there are
745 # no automatic captions but there are subtitles
746 except (KeyError, ExtractorError
):
747 self
._downloader
.report_warning(err_msg
)
751 def extract_id(cls
, url
):
752 mobj
= re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
)
754 raise ExtractorError('Invalid URL: %s' % url
)
755 video_id
= mobj
.group(2)
758 def _extract_from_m3u8(self
, manifest_url
, video_id
):
761 def _get_urls(_manifest
):
762 lines
= _manifest
.split('\n')
763 urls
= filter(lambda l
: l
and not l
.startswith('#'),
766 manifest
= self
._download
_webpage
(manifest_url
, video_id
, 'Downloading formats manifest')
767 formats_urls
= _get_urls(manifest
)
768 for format_url
in formats_urls
:
769 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
770 url_map
[itag
] = format_url
773 def _extract_annotations(self
, video_id
):
774 url
= 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
775 return self
._download
_webpage
(url
, video_id
, note
='Searching for annotations.', errnote
='Unable to download video annotations.')
777 def _parse_dash_manifest(
778 self
, video_id
, dash_manifest_url
, player_url
, age_gate
):
779 def decrypt_sig(mobj
):
781 dec_s
= self
._decrypt
_signature
(s
, video_id
, player_url
, age_gate
)
782 return '/signature/%s' % dec_s
783 dash_manifest_url
= re
.sub(r
'/s/([\w\.]+)', decrypt_sig
, dash_manifest_url
)
784 dash_doc
= self
._download
_xml
(
785 dash_manifest_url
, video_id
,
786 note
='Downloading DASH manifest',
787 errnote
='Could not download DASH manifest')
790 for a
in dash_doc
.findall('.//{urn:mpeg:DASH:schema:MPD:2011}AdaptationSet'):
791 mime_type
= a
.attrib
.get('mimeType')
792 for r
in a
.findall('{urn:mpeg:DASH:schema:MPD:2011}Representation'):
793 url_el
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
796 if mime_type
== 'text/vtt':
797 # TODO implement WebVTT downloading
799 elif mime_type
.startswith('audio/') or mime_type
.startswith('video/'):
800 format_id
= r
.attrib
['id']
801 video_url
= url_el
.text
802 filesize
= int_or_none(url_el
.attrib
.get('{http://youtube.com/yt/2012/10/10}contentLength'))
804 'format_id': format_id
,
806 'width': int_or_none(r
.attrib
.get('width')),
807 'height': int_or_none(r
.attrib
.get('height')),
808 'tbr': int_or_none(r
.attrib
.get('bandwidth'), 1000),
809 'asr': int_or_none(r
.attrib
.get('audioSamplingRate')),
810 'filesize': filesize
,
811 'fps': int_or_none(r
.attrib
.get('frameRate')),
814 existing_format
= next(
816 if fo
['format_id'] == format_id
)
817 except StopIteration:
818 full_info
= self
._formats
.get(format_id
, {}).copy()
820 formats
.append(full_info
)
822 existing_format
.update(f
)
824 self
.report_warning('Unknown MIME type %s in DASH manifest' % mime_type
)
827 def _real_extract(self
, url
):
829 'http' if self
._downloader
.params
.get('prefer_insecure', False)
832 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
833 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
835 url
= proto
+ '://www.youtube.com/' + compat_urllib_parse
.unquote(mobj
.group(1)).lstrip('/')
836 video_id
= self
.extract_id(url
)
839 url
= proto
+ '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
840 video_webpage
= self
._download
_webpage
(url
, video_id
)
842 # Attempt to extract SWF player URL
843 mobj
= re
.search(r
'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
845 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
851 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
853 # We simulate the access to the video from www.youtube.com/v/{video_id}
854 # this can be viewed without login into Youtube
855 url
= proto
+ '://www.youtube.com/embed/%s' % video_id
856 embed_webpage
= self
._download
_webpage
(url
, video_id
, 'Downloading embed webpage')
857 data
= compat_urllib_parse
.urlencode({
858 'video_id': video_id
,
859 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
860 'sts': self
._search
_regex
(
861 r
'"sts"\s*:\s*(\d+)', embed_webpage
, 'sts', default
=''),
863 video_info_url
= proto
+ '://www.youtube.com/get_video_info?' + data
864 video_info_webpage
= self
._download
_webpage
(
865 video_info_url
, video_id
,
866 note
='Refetching age-gated info webpage',
867 errnote
='unable to download video info webpage')
868 video_info
= compat_parse_qs(video_info_webpage
)
872 # Try looking directly into the video webpage
873 mobj
= re
.search(r
';ytplayer\.config\s*=\s*({.*?});', video_webpage
)
875 raise ValueError('Could not find ytplayer.config') # caught below
876 json_code
= uppercase_escape(mobj
.group(1))
877 ytplayer_config
= json
.loads(json_code
)
878 args
= ytplayer_config
['args']
879 # Convert to the same format returned by compat_parse_qs
880 video_info
= dict((k
, [v
]) for k
, v
in args
.items())
881 if not args
.get('url_encoded_fmt_stream_map'):
882 raise ValueError('No stream_map present') # caught below
884 # We fallback to the get_video_info pages (used by the embed page)
885 self
.report_video_info_webpage_download(video_id
)
886 for el_type
in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
888 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
889 % (proto
, video_id
, el_type
))
890 video_info_webpage
= self
._download
_webpage
(
892 video_id
, note
=False,
893 errnote
='unable to download video info webpage')
894 video_info
= compat_parse_qs(video_info_webpage
)
895 if 'token' in video_info
:
897 if 'token' not in video_info
:
898 if 'reason' in video_info
:
899 raise ExtractorError(
900 'YouTube said: %s' % video_info
['reason'][0],
901 expected
=True, video_id
=video_id
)
903 raise ExtractorError(
904 '"token" parameter not in video info for unknown reason',
907 if 'view_count' in video_info
:
908 view_count
= int(video_info
['view_count'][0])
912 # Check for "rental" videos
913 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
914 raise ExtractorError('"rental" videos not supported')
916 # Start extracting information
917 self
.report_information_extraction(video_id
)
920 if 'author' not in video_info
:
921 raise ExtractorError('Unable to extract uploader name')
922 video_uploader
= compat_urllib_parse
.unquote_plus(video_info
['author'][0])
925 video_uploader_id
= None
926 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
928 video_uploader_id
= mobj
.group(1)
930 self
._downloader
.report_warning('unable to extract uploader nickname')
933 if 'title' in video_info
:
934 video_title
= video_info
['title'][0]
936 self
._downloader
.report_warning('Unable to extract video title')
940 # We try first to get a high quality image:
941 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
942 video_webpage
, re
.DOTALL
)
943 if m_thumb
is not None:
944 video_thumbnail
= m_thumb
.group(1)
945 elif 'thumbnail_url' not in video_info
:
946 self
._downloader
.report_warning('unable to extract video thumbnail')
947 video_thumbnail
= None
948 else: # don't panic if we can't find it
949 video_thumbnail
= compat_urllib_parse
.unquote_plus(video_info
['thumbnail_url'][0])
953 mobj
= re
.search(r
'(?s)id="eow-date.*?>(.*?)</span>', video_webpage
)
956 r
'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
959 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
960 upload_date
= unified_strdate(upload_date
)
962 m_cat_container
= self
._search
_regex
(
963 r
'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
964 video_webpage
, 'categories', default
=None)
966 category
= self
._html
_search
_regex
(
967 r
'(?s)<a[^<]+>(.*?)</a>', m_cat_container
, 'category',
969 video_categories
= None if category
is None else [category
]
971 video_categories
= None
974 video_description
= get_element_by_id("eow-description", video_webpage
)
975 if video_description
:
976 video_description
= re
.sub(r
'''(?x)
978 (?:[a-zA-Z-]+="[^"]+"\s+)*?
980 (?:[a-zA-Z-]+="[^"]+"\s+)*?
981 class="yt-uix-redirect-link"\s*>
984 ''', r
'\1', video_description
)
985 video_description
= clean_html(video_description
)
987 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
989 video_description
= unescapeHTML(fd_mobj
.group(1))
991 video_description
= ''
993 def _extract_count(count_name
):
994 count
= self
._search
_regex
(
995 r
'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re
.escape(count_name
),
996 video_webpage
, count_name
, default
=None)
997 if count
is not None:
998 return int(count
.replace(',', ''))
1000 like_count
= _extract_count('like')
1001 dislike_count
= _extract_count('dislike')
1004 video_subtitles
= self
.extract_subtitles(video_id
, video_webpage
)
1005 automatic_captions
= self
.extract_automatic_captions(video_id
, video_webpage
)
1007 if 'length_seconds' not in video_info
:
1008 self
._downloader
.report_warning('unable to extract video duration')
1009 video_duration
= None
1011 video_duration
= int(compat_urllib_parse
.unquote_plus(video_info
['length_seconds'][0]))
1014 video_annotations
= None
1015 if self
._downloader
.params
.get('writeannotations', False):
1016 video_annotations
= self
._extract
_annotations
(video_id
)
1018 def _map_to_format_list(urlmap
):
1020 for itag
, video_real_url
in urlmap
.items():
1023 'url': video_real_url
,
1024 'player_url': player_url
,
1026 if itag
in self
._formats
:
1027 dct
.update(self
._formats
[itag
])
1031 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
1032 self
.report_rtmp_download()
1034 'format_id': '_rtmp',
1036 'url': video_info
['conn'][0],
1037 'player_url': player_url
,
1039 elif len(video_info
.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info
.get('adaptive_fmts', [''])[0]) >= 1:
1040 encoded_url_map
= video_info
.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info
.get('adaptive_fmts', [''])[0]
1041 if 'rtmpe%3Dyes' in encoded_url_map
:
1042 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
1044 for url_data_str
in encoded_url_map
.split(','):
1045 url_data
= compat_parse_qs(url_data_str
)
1046 if 'itag' not in url_data
or 'url' not in url_data
:
1048 format_id
= url_data
['itag'][0]
1049 url
= url_data
['url'][0]
1051 if 'sig' in url_data
:
1052 url
+= '&signature=' + url_data
['sig'][0]
1053 elif 's' in url_data
:
1054 encrypted_sig
= url_data
['s'][0]
1055 ASSETS_RE
= r
'"assets":.+?"js":\s*("[^"]+")'
1057 jsplayer_url_json
= self
._search
_regex
(
1059 embed_webpage
if age_gate
else video_webpage
,
1060 'JS player URL (1)', default
=None)
1061 if not jsplayer_url_json
and not age_gate
:
1062 # We need the embed website after all
1063 if embed_webpage
is None:
1064 embed_url
= proto
+ '://www.youtube.com/embed/%s' % video_id
1065 embed_webpage
= self
._download
_webpage
(
1066 embed_url
, video_id
, 'Downloading embed webpage')
1067 jsplayer_url_json
= self
._search
_regex
(
1068 ASSETS_RE
, embed_webpage
, 'JS player URL')
1070 player_url
= json
.loads(jsplayer_url_json
)
1071 if player_url
is None:
1072 player_url_json
= self
._search
_regex
(
1073 r
'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1074 video_webpage
, 'age gate player URL')
1075 player_url
= json
.loads(player_url_json
)
1077 if self
._downloader
.params
.get('verbose'):
1078 if player_url
is None:
1079 player_version
= 'unknown'
1080 player_desc
= 'unknown'
1082 if player_url
.endswith('swf'):
1083 player_version
= self
._search
_regex
(
1084 r
'-(.+?)(?:/watch_as3)?\.swf$', player_url
,
1085 'flash player', fatal
=False)
1086 player_desc
= 'flash player %s' % player_version
1088 player_version
= self
._search
_regex
(
1089 r
'html5player-([^/]+?)(?:/html5player)?\.js',
1091 'html5 player', fatal
=False)
1092 player_desc
= 'html5 player %s' % player_version
1094 parts_sizes
= self
._signature
_cache
_id
(encrypted_sig
)
1095 self
.to_screen('{%s} signature length %s, %s' %
1096 (format_id
, parts_sizes
, player_desc
))
1098 signature
= self
._decrypt
_signature
(
1099 encrypted_sig
, video_id
, player_url
, age_gate
)
1100 url
+= '&signature=' + signature
1101 if 'ratebypass' not in url
:
1102 url
+= '&ratebypass=yes'
1103 url_map
[format_id
] = url
1104 formats
= _map_to_format_list(url_map
)
1105 elif video_info
.get('hlsvp'):
1106 manifest_url
= video_info
['hlsvp'][0]
1107 url_map
= self
._extract
_from
_m
3u8(manifest_url
, video_id
)
1108 formats
= _map_to_format_list(url_map
)
1110 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1112 # Look for the DASH manifest
1113 if self
._downloader
.params
.get('youtube_include_dash_manifest', True):
1114 dash_mpd
= video_info
.get('dashmpd')
1116 dash_manifest_url
= dash_mpd
[0]
1118 dash_formats
= self
._parse
_dash
_manifest
(
1119 video_id
, dash_manifest_url
, player_url
, age_gate
)
1120 except (ExtractorError
, KeyError) as e
:
1121 self
.report_warning(
1122 'Skipping DASH manifest: %r' % e
, video_id
)
1124 # Hide the formats we found through non-DASH
1125 dash_keys
= set(df
['format_id'] for df
in dash_formats
)
1127 if f
['format_id'] in dash_keys
:
1128 f
['format_id'] = 'nondash-%s' % f
['format_id']
1129 f
['preference'] = f
.get('preference', 0) - 10000
1130 formats
.extend(dash_formats
)
1132 # Check for malformed aspect ratio
1133 stretched_m
= re
.search(
1134 r
'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
1137 ratio
= float(stretched_m
.group('w')) / float(stretched_m
.group('h'))
1139 if f
.get('vcodec') != 'none':
1140 f
['stretched_ratio'] = ratio
1142 self
._sort
_formats
(formats
)
1146 'uploader': video_uploader
,
1147 'uploader_id': video_uploader_id
,
1148 'upload_date': upload_date
,
1149 'title': video_title
,
1150 'thumbnail': video_thumbnail
,
1151 'description': video_description
,
1152 'categories': video_categories
,
1153 'subtitles': video_subtitles
,
1154 'automatic_captions': automatic_captions
,
1155 'duration': video_duration
,
1156 'age_limit': 18 if age_gate
else 0,
1157 'annotations': video_annotations
,
1158 'webpage_url': proto
+ '://www.youtube.com/watch?v=%s' % video_id
,
1159 'view_count': view_count
,
1160 'like_count': like_count
,
1161 'dislike_count': dislike_count
,
1162 'average_rating': float_or_none(video_info
.get('avg_rating', [None])[0]),
1167 class YoutubePlaylistIE(YoutubeBaseInfoExtractor
):
1168 IE_DESC
= 'YouTube.com playlists'
1169 _VALID_URL
= r
"""(?x)(?:
1174 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1175 \? (?:.*?&)*? (?:p|a|list)=
1179 (?:PL|LL|EC|UU|FL|RD|UL)?[0-9A-Za-z-_]{10,}
1180 # Top tracks, they can also include dots
1185 ((?:PL|LL|EC|UU|FL|RD|UL)[0-9A-Za-z-_]{10,})
1187 _TEMPLATE_URL
= 'https://www.youtube.com/playlist?list=%s'
1188 _VIDEO_RE
= r
'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1189 IE_NAME
= 'youtube:playlist'
1191 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1193 'title': 'ytdl test PL',
1194 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1196 'playlist_count': 3,
1198 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1200 'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1201 'title': 'YDL_Empty_List',
1203 'playlist_count': 0,
1205 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1206 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1208 'title': '29C3: Not my department',
1209 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1211 'playlist_count': 95,
1213 'note': 'issue #673',
1214 'url': 'PLBB231211A4F62143',
1216 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1217 'id': 'PLBB231211A4F62143',
1219 'playlist_mincount': 26,
1221 'note': 'Large playlist',
1222 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1224 'title': 'Uploads from Cauchemar',
1225 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
1227 'playlist_mincount': 799,
1229 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1231 'title': 'YDL_safe_search',
1232 'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1234 'playlist_count': 2,
1237 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1238 'playlist_count': 4,
1241 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1244 'note': 'Embedded SWF player',
1245 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1246 'playlist_count': 4,
1249 'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
1252 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1253 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1255 'title': 'Uploads from Interstellar Movie',
1256 'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
1258 'playlist_mincout': 21,
1261 def _real_initialize(self
):
1264 def _ids_to_results(self
, ids
):
1266 self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
1269 def _extract_mix(self
, playlist_id
):
1270 # The mixes are generated from a single video
1271 # the id of the playlist is just 'RD' + video_id
1272 url
= 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id
[-11:], playlist_id
)
1273 webpage
= self
._download
_webpage
(
1274 url
, playlist_id
, 'Downloading Youtube mix')
1275 search_title
= lambda class_name
: get_element_by_attribute('class', class_name
, webpage
)
1277 search_title('playlist-title') or
1278 search_title('title long-title') or
1279 search_title('title'))
1280 title
= clean_html(title_span
)
1281 ids
= orderedSet(re
.findall(
1282 r
'''(?xs)data-video-username=".*?".*?
1283 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re
.escape(playlist_id
),
1285 url_results
= self
._ids
_to
_results
(ids
)
1287 return self
.playlist_result(url_results
, playlist_id
, title
)
1289 def _extract_playlist(self
, playlist_id
):
1290 url
= self
._TEMPLATE
_URL
% playlist_id
1291 page
= self
._download
_webpage
(url
, playlist_id
)
1292 more_widget_html
= content_html
= page
1294 # Check if the playlist exists or is private
1295 if re
.search(r
'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page
) is not None:
1296 raise ExtractorError(
1297 'The playlist doesn\'t exist or is private, use --username or '
1298 '--netrc to access it.',
1301 # Extract the video ids from the playlist pages
1304 for page_num
in itertools
.count(1):
1305 matches
= re
.finditer(self
._VIDEO
_RE
, content_html
)
1306 # We remove the duplicates and the link with index 0
1307 # (it's not the first video of the playlist)
1308 new_ids
= orderedSet(m
.group('id') for m
in matches
if m
.group('index') != '0')
1311 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1315 more
= self
._download
_json
(
1316 'https://youtube.com/%s' % mobj
.group('more'), playlist_id
,
1317 'Downloading page #%s' % page_num
,
1318 transform_source
=uppercase_escape
)
1319 content_html
= more
['content_html']
1320 if not content_html
.strip():
1321 # Some webpages show a "Load more" button but they don't
1324 more_widget_html
= more
['load_more_widget_html']
1326 playlist_title
= self
._html
_search
_regex
(
1327 r
'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1330 url_results
= self
._ids
_to
_results
(ids
)
1331 return self
.playlist_result(url_results
, playlist_id
, playlist_title
)
1333 def _real_extract(self
, url
):
1334 # Extract playlist id
1335 mobj
= re
.match(self
._VALID
_URL
, url
)
1337 raise ExtractorError('Invalid URL: %s' % url
)
1338 playlist_id
= mobj
.group(1) or mobj
.group(2)
1340 # Check if it's a video-specific URL
1341 query_dict
= compat_urlparse
.parse_qs(compat_urlparse
.urlparse(url
).query
)
1342 if 'v' in query_dict
:
1343 video_id
= query_dict
['v'][0]
1344 if self
._downloader
.params
.get('noplaylist'):
1345 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1346 return self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1348 self
.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id
, video_id
))
1350 if playlist_id
.startswith('RD') or playlist_id
.startswith('UL'):
1351 # Mixes require a custom extraction process
1352 return self
._extract
_mix
(playlist_id
)
1354 return self
._extract
_playlist
(playlist_id
)
1357 class YoutubeChannelIE(InfoExtractor
):
1358 IE_DESC
= 'YouTube.com channels'
1359 _VALID_URL
= r
'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1360 _TEMPLATE_URL
= 'https://www.youtube.com/channel/%s/videos'
1361 IE_NAME
= 'youtube:channel'
1363 'note': 'paginated channel',
1364 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1365 'playlist_mincount': 91,
1367 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w',
1372 def extract_videos_from_page(page
):
1375 for mobj
in re
.finditer(r
'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?', page
):
1376 video_id
= mobj
.group('id')
1377 video_title
= unescapeHTML(mobj
.group('title'))
1379 idx
= ids_in_page
.index(video_id
)
1380 if video_title
and not titles_in_page
[idx
]:
1381 titles_in_page
[idx
] = video_title
1383 ids_in_page
.append(video_id
)
1384 titles_in_page
.append(video_title
)
1385 return zip(ids_in_page
, titles_in_page
)
1387 def _real_extract(self
, url
):
1388 channel_id
= self
._match
_id
(url
)
1390 url
= self
._TEMPLATE
_URL
% channel_id
1391 channel_page
= self
._download
_webpage
(url
, channel_id
, 'Downloading page #1')
1392 autogenerated
= re
.search(r
'''(?x)
1394 channel-header-autogenerated-label|
1395 yt-channel-title-autogenerated
1396 )[^"]*"''', channel_page
) is not None
1399 # The videos are contained in a single page
1400 # the ajax pages can't be used, they are empty
1403 video_id
, 'Youtube', video_id
=video_id
,
1404 video_title
=video_title
)
1405 for video_id
, video_title
in self
.extract_videos_from_page(channel_page
)]
1406 return self
.playlist_result(entries
, channel_id
)
1409 more_widget_html
= content_html
= channel_page
1410 for pagenum
in itertools
.count(1):
1412 for video_id
, video_title
in self
.extract_videos_from_page(content_html
):
1413 yield self
.url_result(
1414 video_id
, 'Youtube', video_id
=video_id
,
1415 video_title
=video_title
)
1418 r
'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1423 more
= self
._download
_json
(
1424 'https://youtube.com/%s' % mobj
.group('more'), channel_id
,
1425 'Downloading page #%s' % (pagenum
+ 1),
1426 transform_source
=uppercase_escape
)
1427 content_html
= more
['content_html']
1428 more_widget_html
= more
['load_more_widget_html']
1430 return self
.playlist_result(_entries(), channel_id
)
1433 class YoutubeUserIE(YoutubeChannelIE
):
1434 IE_DESC
= 'YouTube.com user videos (URL or "ytuser" keyword)'
1435 _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_-]+)'
1436 _TEMPLATE_URL
= 'https://www.youtube.com/user/%s/videos'
1437 IE_NAME
= 'youtube:user'
1440 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1441 'playlist_mincount': 320,
1443 'title': 'TheLinuxFoundation',
1446 'url': 'ytuser:phihag',
1447 'only_matching': True,
1451 def suitable(cls
, url
):
1452 # Don't return True if the url can be extracted with other youtube
1453 # extractor, the regex would is too permissive and it would match.
1454 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1455 if any(ie
.suitable(url
) for ie
in other_ies
):
1458 return super(YoutubeUserIE
, cls
).suitable(url
)
1461 class YoutubeSearchIE(SearchInfoExtractor
, YoutubePlaylistIE
):
1462 IE_DESC
= 'YouTube.com searches'
1463 # there doesn't appear to be a real limit, for example if you search for
1464 # 'python' you get more than 8.000.000 results
1465 _MAX_RESULTS
= float('inf')
1466 IE_NAME
= 'youtube:search'
1467 _SEARCH_KEY
= 'ytsearch'
1468 _EXTRA_QUERY_ARGS
= {}
1470 def _get_n_results(self
, query
, n
):
1471 """Get a specified number of results for a query"""
1476 for pagenum
in itertools
.count(1):
1478 'search_query': query
,
1482 url_query
.update(self
._EXTRA
_QUERY
_ARGS
)
1483 result_url
= 'https://www.youtube.com/results?' + compat_urllib_parse
.urlencode(url_query
)
1484 data
= self
._download
_json
(
1485 result_url
, video_id
='query "%s"' % query
,
1486 note
='Downloading page %s' % pagenum
,
1487 errnote
='Unable to download API page')
1488 html_content
= data
[1]['body']['content']
1490 if 'class="search-message' in html_content
:
1491 raise ExtractorError(
1492 '[youtube] No video results', expected
=True)
1494 new_videos
= self
._ids
_to
_results
(orderedSet(re
.findall(
1495 r
'href="/watch\?v=(.{11})', html_content
)))
1496 videos
+= new_videos
1497 if not new_videos
or len(videos
) > limit
:
1502 return self
.playlist_result(videos
, query
)
1505 class YoutubeSearchDateIE(YoutubeSearchIE
):
1506 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1507 _SEARCH_KEY
= 'ytsearchdate'
1508 IE_DESC
= 'YouTube.com searches, newest videos first'
1509 _EXTRA_QUERY_ARGS
= {'search_sort': 'video_date_uploaded'}
1512 class YoutubeSearchURLIE(InfoExtractor
):
1513 IE_DESC
= 'YouTube.com search URLs'
1514 IE_NAME
= 'youtube:search_url'
1515 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1517 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1518 'playlist_mincount': 5,
1520 'title': 'youtube-dl test video',
1524 def _real_extract(self
, url
):
1525 mobj
= re
.match(self
._VALID
_URL
, url
)
1526 query
= compat_urllib_parse
.unquote_plus(mobj
.group('query'))
1528 webpage
= self
._download
_webpage
(url
, query
)
1529 result_code
= self
._search
_regex
(
1530 r
'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage
, 'result HTML')
1532 part_codes
= re
.findall(
1533 r
'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code
)
1535 for part_code
in part_codes
:
1536 part_title
= self
._html
_search
_regex
(
1537 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1538 part_url_snippet
= self
._html
_search
_regex
(
1539 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1540 part_url
= compat_urlparse
.urljoin(
1541 'https://www.youtube.com/', part_url_snippet
)
1545 'title': part_title
,
1549 '_type': 'playlist',
1555 class YoutubeShowIE(InfoExtractor
):
1556 IE_DESC
= 'YouTube.com (multi-season) shows'
1557 _VALID_URL
= r
'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1558 IE_NAME
= 'youtube:show'
1560 'url': 'http://www.youtube.com/show/airdisasters',
1561 'playlist_mincount': 3,
1563 'id': 'airdisasters',
1564 'title': 'Air Disasters',
1568 def _real_extract(self
, url
):
1569 mobj
= re
.match(self
._VALID
_URL
, url
)
1570 playlist_id
= mobj
.group('id')
1571 webpage
= self
._download
_webpage
(
1572 url
, playlist_id
, 'Downloading show webpage')
1573 # There's one playlist for each season of the show
1574 m_seasons
= list(re
.finditer(r
'href="(/playlist\?list=.*?)"', webpage
))
1575 self
.to_screen('%s: Found %s seasons' % (playlist_id
, len(m_seasons
)))
1578 'https://www.youtube.com' + season
.group(1), 'YoutubePlaylist')
1579 for season
in m_seasons
1581 title
= self
._og
_search
_title
(webpage
, fatal
=False)
1584 '_type': 'playlist',
1591 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1593 Base class for extractors that fetch info from
1594 http://www.youtube.com/feed_ajax
1595 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1597 _LOGIN_REQUIRED
= True
1598 # use action_load_personal_feed instead of action_load_system_feed
1599 _PERSONAL_FEED
= False
1602 def _FEED_TEMPLATE(self
):
1603 action
= 'action_load_system_feed'
1604 if self
._PERSONAL
_FEED
:
1605 action
= 'action_load_personal_feed'
1606 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action
, self
._FEED
_NAME
)
1610 return 'youtube:%s' % self
._FEED
_NAME
1612 def _real_initialize(self
):
1615 def _real_extract(self
, url
):
1618 for i
in itertools
.count(1):
1619 info
= self
._download
_json
(
1620 self
._FEED
_TEMPLATE
% paging
,
1621 '%s feed' % self
._FEED
_NAME
,
1622 'Downloading page %s' % i
,
1623 transform_source
=uppercase_escape
)
1624 feed_html
= info
.get('feed_html') or info
.get('content_html')
1625 load_more_widget_html
= info
.get('load_more_widget_html') or feed_html
1626 m_ids
= re
.finditer(r
'"/watch\?v=(.*?)["&]', feed_html
)
1627 ids
= orderedSet(m
.group(1) for m
in m_ids
)
1628 feed_entries
.extend(
1629 self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1630 for video_id
in ids
)
1632 r
'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1633 load_more_widget_html
)
1636 paging
= mobj
.group('paging')
1637 return self
.playlist_result(feed_entries
, playlist_title
=self
._PLAYLIST
_TITLE
)
1640 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
1641 IE_NAME
= 'youtube:recommended'
1642 IE_DESC
= 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1643 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1644 _FEED_NAME
= 'recommended'
1645 _PLAYLIST_TITLE
= 'Youtube Recommended videos'
1648 class YoutubeWatchLaterIE(YoutubePlaylistIE
):
1649 IE_NAME
= 'youtube:watchlater'
1650 IE_DESC
= 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1651 _VALID_URL
= r
'https?://www\.youtube\.com/(?:feed/watch_later|playlist\?list=WL)|:ytwatchlater'
1653 _TESTS
= [] # override PlaylistIE tests
1655 def _real_extract(self
, url
):
1656 return self
._extract
_playlist
('WL')
1659 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
1660 IE_NAME
= 'youtube:history'
1661 IE_DESC
= 'Youtube watch history, ":ythistory" for short (requires authentication)'
1662 _VALID_URL
= 'https?://www\.youtube\.com/feed/history|:ythistory'
1663 _FEED_NAME
= 'history'
1664 _PERSONAL_FEED
= True
1665 _PLAYLIST_TITLE
= 'Youtube Watch History'
1668 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
1669 IE_NAME
= 'youtube:favorites'
1670 IE_DESC
= 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1671 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1672 _LOGIN_REQUIRED
= True
1674 def _real_extract(self
, url
):
1675 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1676 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, 'favourites playlist id')
1677 return self
.url_result(playlist_id
, 'YoutubePlaylist')
1680 class YoutubeSubscriptionsIE(YoutubePlaylistIE
):
1681 IE_NAME
= 'youtube:subscriptions'
1682 IE_DESC
= 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1683 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1686 def _real_extract(self
, url
):
1687 title
= 'Youtube Subscriptions'
1688 page
= self
._download
_webpage
('https://www.youtube.com/feed/subscriptions', title
)
1690 # The extraction process is the same as for playlists, but the regex
1691 # for the video ids doesn't contain an index
1693 more_widget_html
= content_html
= page
1695 for page_num
in itertools
.count(1):
1696 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
1697 new_ids
= orderedSet(matches
)
1700 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1704 more
= self
._download
_json
(
1705 'https://youtube.com/%s' % mobj
.group('more'), title
,
1706 'Downloading page #%s' % page_num
,
1707 transform_source
=uppercase_escape
)
1708 content_html
= more
['content_html']
1709 more_widget_html
= more
['load_more_widget_html']
1712 '_type': 'playlist',
1714 'entries': self
._ids
_to
_results
(ids
),
1718 class YoutubeTruncatedURLIE(InfoExtractor
):
1719 IE_NAME
= 'youtube:truncated_url'
1720 IE_DESC
= False # Do not list
1721 _VALID_URL
= r
'''(?x)
1723 (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
1726 annotation_id=annotation_[^&]+|
1731 attribution_link\?a=[^&]+
1737 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1738 'only_matching': True,
1740 'url': 'http://www.youtube.com/watch?',
1741 'only_matching': True,
1743 'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
1744 'only_matching': True,
1746 'url': 'https://www.youtube.com/watch?feature=foo',
1747 'only_matching': True,
1749 'url': 'https://www.youtube.com/watch?hl=en-GB',
1750 'only_matching': True,
1753 def _real_extract(self
, url
):
1754 raise ExtractorError(
1755 'Did you forget to quote the URL? Remember that & is a meta '
1756 'character in most shells, so you want to put the URL in quotes, '
1758 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1759 ' or simply youtube-dl BaW_jenozKc .',
1763 class YoutubeTruncatedIDIE(InfoExtractor
):
1764 IE_NAME
= 'youtube:truncated_id'
1765 IE_DESC
= False # Do not list
1766 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1769 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1770 'only_matching': True,
1773 def _real_extract(self
, url
):
1774 video_id
= self
._match
_id
(url
)
1775 raise ExtractorError(
1776 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id
, url
),