3 from __future__
import unicode_literals
13 from .common
import InfoExtractor
, SearchInfoExtractor
14 from .subtitles
import SubtitlesInfoExtractor
15 from ..jsinterp
import JSInterpreter
16 from ..swfinterp
import SWFInterpreter
17 from ..compat
import (
21 compat_urllib_request
,
28 get_element_by_attribute
,
39 class YoutubeBaseInfoExtractor(InfoExtractor
):
40 """Provide base functions for Youtube extractors"""
41 _LOGIN_URL
= 'https://accounts.google.com/ServiceLogin'
42 _TWOFACTOR_URL
= 'https://accounts.google.com/SecondFactor'
43 _NETRC_MACHINE
= 'youtube'
44 # If True it will raise an error if no login info is provided
45 _LOGIN_REQUIRED
= False
47 def _set_language(self
):
49 '.youtube.com', 'PREF', 'f1=50000000&hl=en',
50 # YouTube sets the expire time to about two months
51 expire_time
=time
.time() + 2 * 30 * 24 * 3600)
55 Attempt to log in to YouTube.
56 True is returned if successful or skipped.
57 False is returned if login failed.
59 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
61 (username
, password
) = self
._get
_login
_info
()
62 # No authentication to be performed
64 if self
._LOGIN
_REQUIRED
:
65 raise ExtractorError('No login info available, needed for using %s.' % self
.IE_NAME
, expected
=True)
68 login_page
= self
._download
_webpage
(
69 self
._LOGIN
_URL
, None,
70 note
='Downloading login page',
71 errnote
='unable to fetch login page', fatal
=False)
72 if login_page
is False:
75 galx
= self
._search
_regex
(r
'(?s)<input.+?name="GALX".+?value="(.+?)"',
76 login_page
, 'Login GALX parameter')
80 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
85 'PersistentCookie': 'yes',
87 'bgresponse': 'js_disabled',
88 'checkConnection': '',
89 'checkedDomains': 'youtube',
101 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
103 login_form
= dict((k
.encode('utf-8'), v
.encode('utf-8')) for k
, v
in login_form_strs
.items())
104 login_data
= compat_urllib_parse
.urlencode(login_form
).encode('ascii')
106 req
= compat_urllib_request
.Request(self
._LOGIN
_URL
, login_data
)
107 login_results
= self
._download
_webpage
(
109 note
='Logging in', errnote
='unable to log in', fatal
=False)
110 if login_results
is False:
113 if re
.search(r
'id="errormsg_0_Passwd"', login_results
) is not None:
114 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected
=True)
117 # TODO add SMS and phone call support - these require making a request and then prompting the user
119 if re
.search(r
'(?i)<form[^>]* id="gaia_secondfactorform"', login_results
) is not None:
120 tfa_code
= self
._get
_tfa
_info
()
123 self
._downloader
.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
124 self
._downloader
.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
127 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
129 match
= re
.search(r
'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
131 self._downloader.report_warning('Failed to get secTok
- did the page structure change?
')
132 secTok = match.group(1)
133 match = re.search(r'id="timeStmp"\n\s
+value
=\'(.+)\'/>', login_results, re.M | re.U)
135 self._downloader.report_warning('Failed to get timeStmp
- did the page structure change?
')
136 timeStmp = match.group(1)
139 'continue': 'https
://www
.youtube
.com
/signin?action_handle_signin
=true
&feature
=sign_in_button
&hl
=en_US
&nomobiletemp
=1',
141 'smsUserPin
': tfa_code,
142 'smsVerifyPin
': 'Verify
',
144 'PersistentCookie
': 'yes
',
145 'checkConnection
': '',
146 'checkedDomains
': 'youtube
',
149 'timeStmp
': timeStmp,
150 'service
': 'youtube
',
153 tfa_form = dict((k.encode('utf
-8'), v.encode('utf
-8')) for k, v in tfa_form_strs.items())
154 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii
')
156 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
157 tfa_results = self._download_webpage(
159 note='Submitting TFA code
', errnote='unable to submit tfa
', fatal=False)
161 if tfa_results is False:
164 if re.search(r'(?i
)<form
[^
>]* id="gaia_secondfactorform"', tfa_results) is not None:
165 self._downloader.report_warning('Two
-factor code expired
. Please
try again
, or use a one
-use backup code instead
.')
167 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', tfa_results) is not None:
168 self._downloader.report_warning('unable to log
in - did the page structure change?
')
170 if re.search(r'smsauth
-interstitial
-reviewsettings
', tfa_results) is not None:
171 self._downloader.report_warning('Your Google account has a security notice
. Please log
in on your web browser
, resolve the notice
, and try again
.')
174 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', login_results) is not None:
175 self._downloader.report_warning('unable to log
in: bad username
or password
')
179 def _real_initialize(self):
180 if self._downloader is None:
183 if not self._login():
187 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
188 IE_DESC = 'YouTube
.com
'
189 _VALID_URL = r"""(?x)^
191 (?:https?://|//) # http(s):// or protocol-independent URL
192 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
193 (?:www\.)?deturl\.com/www\.youtube\.com/|
194 (?:www\.)?pwnyoutube\.com/|
195 (?:www\.)?yourepeat\.com/|
196 tube\.majestyc\.net/|
197 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
198 (?:.*?\#/)? # handle anchor (#/) redirect urls
199 (?: # the various things that can precede the ID:
200 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
201 |(?: # or the v= param in all its forms
202 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
203 (?:\?|\#!?) # the params delimiter ? or # or #!
204 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
208 |youtu\.be/ # just youtu.be/xxxx
209 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
211 )? # all until now is optional -> you can pass the naked ID
212 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
213 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
214 (?(1).+)? # if we found the ID, everything can follow
216 _NEXT_URL_RE = r'[\?&]next_url
=([^
&]+)'
218 '5': {'ext': 'flv', 'width': 400, 'height': 240},
219 '6': {'ext': 'flv', 'width': 450, 'height': 270},
220 '13': {'ext': '3gp'},
221 '17': {'ext': '3gp', 'width': 176, 'height': 144},
222 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
223 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
224 '34': {'ext': 'flv', 'width': 640, 'height': 360},
225 '35': {'ext': 'flv', 'width': 854, 'height': 480},
226 '36': {'ext': '3gp', 'width': 320, 'height': 240},
227 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
228 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
229 '43': {'ext': 'webm', 'width': 640, 'height': 360},
230 '44': {'ext': 'webm', 'width': 854, 'height': 480},
231 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
232 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
236 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
237 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
238 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
239 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
240 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
241 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
242 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
244 # Apple HTTP Live Streaming
245 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
246 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
247 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
248 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
249 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
250 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
251 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
254 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
255 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
256 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
257 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
258 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
259 '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
260 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
261 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
262 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
263 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
264 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
267 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
268 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
269 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
272 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
273 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
274 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
275 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
276 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
277 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
278 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
279 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
280 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
281 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
282 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
283 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
284 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
285 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
286 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
287 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
288 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
289 '303': {'ext': 'webm', 'height': 1080, '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'},
293 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
294 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
296 # Dash webm audio with opus inside
297 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
298 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
299 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
302 '_rtmp
': {'protocol': 'rtmp'},
308 'url
': 'http
://www
.youtube
.com
/watch?v
=BaW_jenozKc
',
312 'title
': 'youtube
-dl test video
"\'/\\ä↭𝕐',
313 'uploader': 'Philipp Hagemeister',
314 'uploader_id': 'phihag',
315 'upload_date': '20121002',
316 '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 .',
317 'categories
': ['Science
& Technology
'],
319 'dislike_count
': int,
323 'url
': 'http
://www
.youtube
.com
/watch?v
=UxxajLWwzqY
',
324 'note
': 'Test generic use_cipher_signature
video (#897)',
328 'upload_date': '20120506',
329 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
330 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
331 'uploader': 'Icona Pop',
332 'uploader_id': 'IconaPop',
336 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
337 'note': 'Test VEVO video with age protection (#956)',
341 'upload_date': '20130703',
342 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
343 'description': 'md5:64249768eec3bc4276236606ea996373',
344 'uploader': 'justintimberlakeVEVO',
345 'uploader_id': 'justintimberlakeVEVO',
349 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
350 'note': 'Embed-only video (#1746)',
354 'upload_date': '20120608',
355 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
356 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
357 'uploader': 'SET India',
358 'uploader_id': 'setindia'
362 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
363 'note': '256k DASH audio (format 141) via DASH manifest',
367 'upload_date': '20121002',
368 'uploader_id': '8KVIDEO',
370 'uploader': '8KVIDEO',
371 'title': 'UHDTV TEST 8K VIDEO.mp4'
374 'youtube_include_dash_manifest': True,
378 # DASH manifest with encrypted signature
380 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
384 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
385 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
386 'uploader': 'AfrojackVEVO',
387 'uploader_id': 'AfrojackVEVO',
388 'upload_date': '20131011',
391 'youtube_include_dash_manifest': True,
397 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
401 'upload_date': '20100909',
402 'uploader': 'The Amazing Atheist',
403 'uploader_id': 'TheAmazingAtheist',
404 'title': 'Burning Everyone\'s Koran',
405 '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',
408 # Normal age-gate video (No vevo, embed allowed)
410 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
414 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
415 'description': 'md5:eca57043abae25130f58f655ad9a7771',
416 'uploader': 'The Witcher',
417 'uploader_id': 'WitcherGame',
418 'upload_date': '20140605',
421 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
423 'url': '__2ABJjxzNo',
427 'upload_date': '20100430',
428 'uploader_id': 'deadmau5',
429 'description': 'md5:12c56784b8032162bb936a5f76d55360',
430 'uploader': 'deadmau5',
431 'title': 'Deadmau5 - Some Chords (HD)',
433 'expected_warnings': [
434 'DASH manifest missing',
437 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
439 'url': 'lqQg6PlCWgI',
448 def __init__(self
, *args
, **kwargs
):
449 super(YoutubeIE
, self
).__init
__(*args
, **kwargs
)
450 self
._player
_cache
= {}
452 def report_video_info_webpage_download(self
, video_id
):
453 """Report attempt to download video info webpage."""
454 self
.to_screen('%s: Downloading video info webpage' % video_id
)
456 def report_information_extraction(self
, video_id
):
457 """Report attempt to extract video information."""
458 self
.to_screen('%s: Extracting video information' % video_id
)
460 def report_unavailable_format(self
, video_id
, format
):
461 """Report extracted video URL."""
462 self
.to_screen('%s: Format %s not available' % (video_id
, format
))
464 def report_rtmp_download(self
):
465 """Indicate the download will use the RTMP protocol."""
466 self
.to_screen('RTMP download detected')
468 def _signature_cache_id(self
, example_sig
):
469 """ Return a string representation of a signature """
470 return '.'.join(compat_str(len(part
)) for part
in example_sig
.split('.'))
472 def _extract_signature_function(self
, video_id
, player_url
, example_sig
):
474 r
'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
477 raise ExtractorError('Cannot identify player %r' % player_url
)
478 player_type
= id_m
.group('ext')
479 player_id
= id_m
.group('id')
481 # Read from filesystem cache
482 func_id
= '%s_%s_%s' % (
483 player_type
, player_id
, self
._signature
_cache
_id
(example_sig
))
484 assert os
.path
.basename(func_id
) == func_id
486 cache_spec
= self
._downloader
.cache
.load('youtube-sigfuncs', func_id
)
487 if cache_spec
is not None:
488 return lambda s
: ''.join(s
[i
] for i
in cache_spec
)
490 if player_type
== 'js':
491 code
= self
._download
_webpage
(
492 player_url
, video_id
,
493 note
='Downloading %s player %s' % (player_type
, player_id
),
494 errnote
='Download of %s failed' % player_url
)
495 res
= self
._parse
_sig
_js
(code
)
496 elif player_type
== 'swf':
497 urlh
= self
._request
_webpage
(
498 player_url
, video_id
,
499 note
='Downloading %s player %s' % (player_type
, player_id
),
500 errnote
='Download of %s failed' % player_url
)
502 res
= self
._parse
_sig
_swf
(code
)
504 assert False, 'Invalid player type %r' % player_type
506 if cache_spec
is None:
507 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
508 cache_res
= res(test_string
)
509 cache_spec
= [ord(c
) for c
in cache_res
]
511 self
._downloader
.cache
.store('youtube-sigfuncs', func_id
, cache_spec
)
514 def _print_sig_code(self
, func
, example_sig
):
515 def gen_sig_code(idxs
):
516 def _genslice(start
, end
, step
):
517 starts
= '' if start
== 0 else str(start
)
518 ends
= (':%d' % (end
+ step
)) if end
+ step
>= 0 else ':'
519 steps
= '' if step
== 1 else (':%d' % step
)
520 return 's[%s%s%s]' % (starts
, ends
, steps
)
523 start
= '(Never used)' # Quelch pyflakes warnings - start will be
524 # set as soon as step is set
525 for i
, prev
in zip(idxs
[1:], idxs
[:-1]):
529 yield _genslice(start
, prev
, step
)
532 if i
- prev
in [-1, 1]:
541 yield _genslice(start
, i
, step
)
543 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
544 cache_res
= func(test_string
)
545 cache_spec
= [ord(c
) for c
in cache_res
]
546 expr_code
= ' + '.join(gen_sig_code(cache_spec
))
547 signature_id_tuple
= '(%s)' % (
548 ', '.join(compat_str(len(p
)) for p
in example_sig
.split('.')))
549 code
= ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
550 ' return %s\n') % (signature_id_tuple
, expr_code
)
551 self
.to_screen('Extracted signature function:\n' + code
)
553 def _parse_sig_js(self
, jscode
):
554 funcname
= self
._search
_regex
(
555 r
'\.sig\|\|([a-zA-Z0-9]+)\(', jscode
,
556 'Initial JS player signature function name')
558 jsi
= JSInterpreter(jscode
)
559 initial_function
= jsi
.extract_function(funcname
)
560 return lambda s
: initial_function([s
])
562 def _parse_sig_swf(self
, file_contents
):
563 swfi
= SWFInterpreter(file_contents
)
564 TARGET_CLASSNAME
= 'SignatureDecipher'
565 searched_class
= swfi
.extract_class(TARGET_CLASSNAME
)
566 initial_function
= swfi
.extract_function(searched_class
, 'decipher')
567 return lambda s
: initial_function([s
])
569 def _decrypt_signature(self
, s
, video_id
, player_url
, age_gate
=False):
570 """Turn the encrypted s field into a working signature"""
572 if player_url
is None:
573 raise ExtractorError('Cannot decrypt signature without player_url')
575 if player_url
.startswith('//'):
576 player_url
= 'https:' + player_url
578 player_id
= (player_url
, self
._signature
_cache
_id
(s
))
579 if player_id
not in self
._player
_cache
:
580 func
= self
._extract
_signature
_function
(
581 video_id
, player_url
, s
583 self
._player
_cache
[player_id
] = func
584 func
= self
._player
_cache
[player_id
]
585 if self
._downloader
.params
.get('youtube_print_sig_code'):
586 self
._print
_sig
_code
(func
, s
)
588 except Exception as e
:
589 tb
= traceback
.format_exc()
590 raise ExtractorError(
591 'Signature extraction failed: ' + tb
, cause
=e
)
593 def _get_available_subtitles(self
, video_id
, webpage
):
595 sub_list
= self
._download
_webpage
(
596 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
,
597 video_id
, note
=False)
598 except ExtractorError
as err
:
599 self
._downloader
.report_warning('unable to download video subtitles: %s' % compat_str(err
))
601 lang_list
= re
.findall(r
'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list
)
606 if lang
in sub_lang_list
:
608 params
= compat_urllib_parse
.urlencode({
611 'fmt': self
._downloader
.params
.get('subtitlesformat', 'srt'),
612 'name': unescapeHTML(l
[0]).encode('utf-8'),
614 url
= 'https://www.youtube.com/api/timedtext?' + params
615 sub_lang_list
[lang
] = url
616 if not sub_lang_list
:
617 self
._downloader
.report_warning('video doesn\'t have subtitles')
621 def _get_available_automatic_caption(self
, video_id
, webpage
):
622 """We need the webpage for getting the captions url, pass it as an
623 argument to speed up the process."""
624 sub_format
= self
._downloader
.params
.get('subtitlesformat', 'srt')
625 self
.to_screen('%s: Looking for automatic captions' % video_id
)
626 mobj
= re
.search(r
';ytplayer.config = ({.*?});', webpage
)
627 err_msg
= 'Couldn\'t find automatic captions for %s' % video_id
629 self
._downloader
.report_warning(err_msg
)
631 player_config
= json
.loads(mobj
.group(1))
633 args
= player_config
['args']
634 caption_url
= args
['ttsurl']
635 timestamp
= args
['timestamp']
636 # We get the available subtitles
637 list_params
= compat_urllib_parse
.urlencode({
642 list_url
= caption_url
+ '&' + list_params
643 caption_list
= self
._download
_xml
(list_url
, video_id
)
644 original_lang_node
= caption_list
.find('track')
645 if original_lang_node
is None or original_lang_node
.attrib
.get('kind') != 'asr':
646 self
._downloader
.report_warning('Video doesn\'t have automatic captions')
648 original_lang
= original_lang_node
.attrib
['lang_code']
651 for lang_node
in caption_list
.findall('target'):
652 sub_lang
= lang_node
.attrib
['lang_code']
653 params
= compat_urllib_parse
.urlencode({
654 'lang': original_lang
,
660 sub_lang_list
[sub_lang
] = caption_url
+ '&' + params
662 # An extractor error can be raise by the download process if there are
663 # no automatic captions but there are subtitles
664 except (KeyError, ExtractorError
):
665 self
._downloader
.report_warning(err_msg
)
669 def extract_id(cls
, url
):
670 mobj
= re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
)
672 raise ExtractorError('Invalid URL: %s' % url
)
673 video_id
= mobj
.group(2)
676 def _extract_from_m3u8(self
, manifest_url
, video_id
):
679 def _get_urls(_manifest
):
680 lines
= _manifest
.split('\n')
681 urls
= filter(lambda l
: l
and not l
.startswith('#'),
684 manifest
= self
._download
_webpage
(manifest_url
, video_id
, 'Downloading formats manifest')
685 formats_urls
= _get_urls(manifest
)
686 for format_url
in formats_urls
:
687 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
688 url_map
[itag
] = format_url
691 def _extract_annotations(self
, video_id
):
692 url
= 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
693 return self
._download
_webpage
(url
, video_id
, note
='Searching for annotations.', errnote
='Unable to download video annotations.')
695 def _parse_dash_manifest(
696 self
, video_id
, dash_manifest_url
, player_url
, age_gate
):
697 def decrypt_sig(mobj
):
699 dec_s
= self
._decrypt
_signature
(s
, video_id
, player_url
, age_gate
)
700 return '/signature/%s' % dec_s
701 dash_manifest_url
= re
.sub(r
'/s/([\w\.]+)', decrypt_sig
, dash_manifest_url
)
702 dash_doc
= self
._download
_xml
(
703 dash_manifest_url
, video_id
,
704 note
='Downloading DASH manifest',
705 errnote
='Could not download DASH manifest')
708 for r
in dash_doc
.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
709 url_el
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
712 format_id
= r
.attrib
['id']
713 video_url
= url_el
.text
714 filesize
= int_or_none(url_el
.attrib
.get('{http://youtube.com/yt/2012/10/10}contentLength'))
716 'format_id': format_id
,
718 'width': int_or_none(r
.attrib
.get('width')),
719 'tbr': int_or_none(r
.attrib
.get('bandwidth'), 1000),
720 'asr': int_or_none(r
.attrib
.get('audioSamplingRate')),
721 'filesize': filesize
,
722 'fps': int_or_none(r
.attrib
.get('frameRate')),
725 existing_format
= next(
727 if fo
['format_id'] == format_id
)
728 except StopIteration:
729 f
.update(self
._formats
.get(format_id
, {}))
732 existing_format
.update(f
)
735 def _real_extract(self
, url
):
737 'http' if self
._downloader
.params
.get('prefer_insecure', False)
740 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
741 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
743 url
= proto
+ '://www.youtube.com/' + compat_urllib_parse
.unquote(mobj
.group(1)).lstrip('/')
744 video_id
= self
.extract_id(url
)
747 url
= proto
+ '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
748 video_webpage
= self
._download
_webpage
(url
, video_id
)
750 # Attempt to extract SWF player URL
751 mobj
= re
.search(r
'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
753 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
758 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
760 # We simulate the access to the video from www.youtube.com/v/{video_id}
761 # this can be viewed without login into Youtube
762 data
= compat_urllib_parse
.urlencode({
763 'video_id': video_id
,
764 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
765 'sts': self
._search
_regex
(
766 r
'"sts"\s*:\s*(\d+)', video_webpage
, 'sts', default
=''),
768 video_info_url
= proto
+ '://www.youtube.com/get_video_info?' + data
769 video_info_webpage
= self
._download
_webpage
(
770 video_info_url
, video_id
,
771 note
='Refetching age-gated info webpage',
772 errnote
='unable to download video info webpage')
773 video_info
= compat_parse_qs(video_info_webpage
)
777 # Try looking directly into the video webpage
778 mobj
= re
.search(r
';ytplayer\.config\s*=\s*({.*?});', video_webpage
)
780 raise ValueError('Could not find ytplayer.config') # caught below
781 json_code
= uppercase_escape(mobj
.group(1))
782 ytplayer_config
= json
.loads(json_code
)
783 args
= ytplayer_config
['args']
784 # Convert to the same format returned by compat_parse_qs
785 video_info
= dict((k
, [v
]) for k
, v
in args
.items())
786 if 'url_encoded_fmt_stream_map' not in args
:
787 raise ValueError('No stream_map present') # caught below
789 # We fallback to the get_video_info pages (used by the embed page)
790 self
.report_video_info_webpage_download(video_id
)
791 for el_type
in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
793 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
794 % (proto
, video_id
, el_type
))
795 video_info_webpage
= self
._download
_webpage
(
797 video_id
, note
=False,
798 errnote
='unable to download video info webpage')
799 video_info
= compat_parse_qs(video_info_webpage
)
800 if 'token' in video_info
:
802 if 'token' not in video_info
:
803 if 'reason' in video_info
:
804 raise ExtractorError(
805 'YouTube said: %s' % video_info
['reason'][0],
806 expected
=True, video_id
=video_id
)
808 raise ExtractorError(
809 '"token" parameter not in video info for unknown reason',
812 if 'view_count' in video_info
:
813 view_count
= int(video_info
['view_count'][0])
817 # Check for "rental" videos
818 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
819 raise ExtractorError('"rental" videos not supported')
821 # Start extracting information
822 self
.report_information_extraction(video_id
)
825 if 'author' not in video_info
:
826 raise ExtractorError('Unable to extract uploader name')
827 video_uploader
= compat_urllib_parse
.unquote_plus(video_info
['author'][0])
830 video_uploader_id
= None
831 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
833 video_uploader_id
= mobj
.group(1)
835 self
._downloader
.report_warning('unable to extract uploader nickname')
838 if 'title' in video_info
:
839 video_title
= video_info
['title'][0]
841 self
._downloader
.report_warning('Unable to extract video title')
845 # We try first to get a high quality image:
846 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
847 video_webpage
, re
.DOTALL
)
848 if m_thumb
is not None:
849 video_thumbnail
= m_thumb
.group(1)
850 elif 'thumbnail_url' not in video_info
:
851 self
._downloader
.report_warning('unable to extract video thumbnail')
852 video_thumbnail
= None
853 else: # don't panic if we can't find it
854 video_thumbnail
= compat_urllib_parse
.unquote_plus(video_info
['thumbnail_url'][0])
858 mobj
= re
.search(r
'(?s)id="eow-date.*?>(.*?)</span>', video_webpage
)
861 r
'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
864 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
865 upload_date
= unified_strdate(upload_date
)
867 m_cat_container
= self
._search
_regex
(
868 r
'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
869 video_webpage
, 'categories', fatal
=False)
871 category
= self
._html
_search
_regex
(
872 r
'(?s)<a[^<]+>(.*?)</a>', m_cat_container
, 'category',
874 video_categories
= None if category
is None else [category
]
876 video_categories
= None
879 video_description
= get_element_by_id("eow-description", video_webpage
)
880 if video_description
:
881 video_description
= re
.sub(r
'''(?x)
883 (?:[a-zA-Z-]+="[^"]+"\s+)*?
885 (?:[a-zA-Z-]+="[^"]+"\s+)*?
886 class="yt-uix-redirect-link"\s*>
889 ''', r
'\1', video_description
)
890 video_description
= clean_html(video_description
)
892 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
894 video_description
= unescapeHTML(fd_mobj
.group(1))
896 video_description
= ''
898 def _extract_count(count_name
):
899 count
= self
._search
_regex
(
900 r
'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re
.escape(count_name
),
901 video_webpage
, count_name
, default
=None)
902 if count
is not None:
903 return int(count
.replace(',', ''))
905 like_count
= _extract_count('like')
906 dislike_count
= _extract_count('dislike')
909 video_subtitles
= self
.extract_subtitles(video_id
, video_webpage
)
911 if self
._downloader
.params
.get('listsubtitles', False):
912 self
._list
_available
_subtitles
(video_id
, video_webpage
)
915 if 'length_seconds' not in video_info
:
916 self
._downloader
.report_warning('unable to extract video duration')
917 video_duration
= None
919 video_duration
= int(compat_urllib_parse
.unquote_plus(video_info
['length_seconds'][0]))
922 video_annotations
= None
923 if self
._downloader
.params
.get('writeannotations', False):
924 video_annotations
= self
._extract
_annotations
(video_id
)
926 def _map_to_format_list(urlmap
):
928 for itag
, video_real_url
in urlmap
.items():
931 'url': video_real_url
,
932 'player_url': player_url
,
934 if itag
in self
._formats
:
935 dct
.update(self
._formats
[itag
])
939 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
940 self
.report_rtmp_download()
942 'format_id': '_rtmp',
944 'url': video_info
['conn'][0],
945 'player_url': player_url
,
947 elif len(video_info
.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info
.get('adaptive_fmts', [])) >= 1:
948 encoded_url_map
= video_info
.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info
.get('adaptive_fmts', [''])[0]
949 if 'rtmpe%3Dyes' in encoded_url_map
:
950 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
952 for url_data_str
in encoded_url_map
.split(','):
953 url_data
= compat_parse_qs(url_data_str
)
954 if 'itag' not in url_data
or 'url' not in url_data
:
956 format_id
= url_data
['itag'][0]
957 url
= url_data
['url'][0]
959 if 'sig' in url_data
:
960 url
+= '&signature=' + url_data
['sig'][0]
961 elif 's' in url_data
:
962 encrypted_sig
= url_data
['s'][0]
965 jsplayer_url_json
= self
._search
_regex
(
966 r
'"assets":.+?"js":\s*("[^"]+")',
967 video_webpage
, 'JS player URL')
968 player_url
= json
.loads(jsplayer_url_json
)
969 if player_url
is None:
970 player_url_json
= self
._search
_regex
(
971 r
'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
972 video_webpage
, 'age gate player URL')
973 player_url
= json
.loads(player_url_json
)
975 if self
._downloader
.params
.get('verbose'):
976 if player_url
is None:
977 player_version
= 'unknown'
978 player_desc
= 'unknown'
980 if player_url
.endswith('swf'):
981 player_version
= self
._search
_regex
(
982 r
'-(.+?)(?:/watch_as3)?\.swf$', player_url
,
983 'flash player', fatal
=False)
984 player_desc
= 'flash player %s' % player_version
986 player_version
= self
._search
_regex
(
987 r
'html5player-([^/]+?)(?:/html5player)?\.js',
989 'html5 player', fatal
=False)
990 player_desc
= 'html5 player %s' % player_version
992 parts_sizes
= self
._signature
_cache
_id
(encrypted_sig
)
993 self
.to_screen('{%s} signature length %s, %s' %
994 (format_id
, parts_sizes
, player_desc
))
996 signature
= self
._decrypt
_signature
(
997 encrypted_sig
, video_id
, player_url
, age_gate
)
998 url
+= '&signature=' + signature
999 if 'ratebypass' not in url
:
1000 url
+= '&ratebypass=yes'
1001 url_map
[format_id
] = url
1002 formats
= _map_to_format_list(url_map
)
1003 elif video_info
.get('hlsvp'):
1004 manifest_url
= video_info
['hlsvp'][0]
1005 url_map
= self
._extract
_from
_m
3u8(manifest_url
, video_id
)
1006 formats
= _map_to_format_list(url_map
)
1008 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1010 # Look for the DASH manifest
1011 if self
._downloader
.params
.get('youtube_include_dash_manifest', True):
1012 dash_mpd
= video_info
.get('dashmpd')
1014 self
.report_warning('%s: DASH manifest missing' % video_id
)
1016 dash_manifest_url
= dash_mpd
[0]
1018 dash_formats
= self
._parse
_dash
_manifest
(
1019 video_id
, dash_manifest_url
, player_url
, age_gate
)
1020 except (ExtractorError
, KeyError) as e
:
1021 self
.report_warning(
1022 'Skipping DASH manifest: %r' % e
, video_id
)
1024 formats
.extend(dash_formats
)
1026 self
._sort
_formats
(formats
)
1030 'uploader': video_uploader
,
1031 'uploader_id': video_uploader_id
,
1032 'upload_date': upload_date
,
1033 'title': video_title
,
1034 'thumbnail': video_thumbnail
,
1035 'description': video_description
,
1036 'categories': video_categories
,
1037 'subtitles': video_subtitles
,
1038 'duration': video_duration
,
1039 'age_limit': 18 if age_gate
else 0,
1040 'annotations': video_annotations
,
1041 'webpage_url': proto
+ '://www.youtube.com/watch?v=%s' % video_id
,
1042 'view_count': view_count
,
1043 'like_count': like_count
,
1044 'dislike_count': dislike_count
,
1049 class YoutubePlaylistIE(YoutubeBaseInfoExtractor
):
1050 IE_DESC
= 'YouTube.com playlists'
1051 _VALID_URL
= r
"""(?x)(?:
1056 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1057 \? (?:.*?&)*? (?:p|a|list)=
1061 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1062 # Top tracks, they can also include dots
1067 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1069 _TEMPLATE_URL
= 'https://www.youtube.com/playlist?list=%s'
1070 _MORE_PAGES_INDICATOR
= r
'data-link-type="next"'
1071 _VIDEO_RE
= r
'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1072 IE_NAME
= 'youtube:playlist'
1074 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1076 'title': 'ytdl test PL',
1077 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1079 'playlist_count': 3,
1081 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1083 'title': 'YDL_Empty_List',
1085 'playlist_count': 0,
1087 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1088 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1090 'title': '29C3: Not my department',
1092 'playlist_count': 95,
1094 'note': 'issue #673',
1095 'url': 'PLBB231211A4F62143',
1097 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1099 'playlist_mincount': 26,
1101 'note': 'Large playlist',
1102 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1104 'title': 'Uploads from Cauchemar',
1106 'playlist_mincount': 799,
1108 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1110 'title': 'YDL_safe_search',
1112 'playlist_count': 2,
1115 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1116 'playlist_count': 4,
1121 'note': 'Embedded SWF player',
1122 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1123 'playlist_count': 4,
1129 def _real_initialize(self
):
1132 def _ids_to_results(self
, ids
):
1134 self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
1137 def _extract_mix(self
, playlist_id
):
1138 # The mixes are generated from a a single video
1139 # the id of the playlist is just 'RD' + video_id
1140 url
= 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id
[-11:], playlist_id
)
1141 webpage
= self
._download
_webpage
(
1142 url
, playlist_id
, 'Downloading Youtube mix')
1143 search_title
= lambda class_name
: get_element_by_attribute('class', class_name
, webpage
)
1145 search_title('playlist-title') or
1146 search_title('title long-title') or
1147 search_title('title'))
1148 title
= clean_html(title_span
)
1149 ids
= orderedSet(re
.findall(
1150 r
'''(?xs)data-video-username=".*?".*?
1151 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re
.escape(playlist_id
),
1153 url_results
= self
._ids
_to
_results
(ids
)
1155 return self
.playlist_result(url_results
, playlist_id
, title
)
1157 def _real_extract(self
, url
):
1158 # Extract playlist id
1159 mobj
= re
.match(self
._VALID
_URL
, url
)
1161 raise ExtractorError('Invalid URL: %s' % url
)
1162 playlist_id
= mobj
.group(1) or mobj
.group(2)
1164 # Check if it's a video-specific URL
1165 query_dict
= compat_urlparse
.parse_qs(compat_urlparse
.urlparse(url
).query
)
1166 if 'v' in query_dict
:
1167 video_id
= query_dict
['v'][0]
1168 if self
._downloader
.params
.get('noplaylist'):
1169 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1170 return self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1172 self
.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id
, video_id
))
1174 if playlist_id
.startswith('RD'):
1175 # Mixes require a custom extraction process
1176 return self
._extract
_mix
(playlist_id
)
1177 if playlist_id
.startswith('TL'):
1178 raise ExtractorError('For downloading YouTube.com top lists, use '
1179 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected
=True)
1181 url
= self
._TEMPLATE
_URL
% playlist_id
1182 page
= self
._download
_webpage
(url
, playlist_id
)
1183 more_widget_html
= content_html
= page
1185 # Check if the playlist exists or is private
1186 if re
.search(r
'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page
) is not None:
1187 raise ExtractorError(
1188 'The playlist doesn\'t exist or is private, use --username or '
1189 '--netrc to access it.',
1192 # Extract the video ids from the playlist pages
1195 for page_num
in itertools
.count(1):
1196 matches
= re
.finditer(self
._VIDEO
_RE
, content_html
)
1197 # We remove the duplicates and the link with index 0
1198 # (it's not the first video of the playlist)
1199 new_ids
= orderedSet(m
.group('id') for m
in matches
if m
.group('index') != '0')
1202 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1206 more
= self
._download
_json
(
1207 'https://youtube.com/%s' % mobj
.group('more'), playlist_id
,
1208 'Downloading page #%s' % page_num
,
1209 transform_source
=uppercase_escape
)
1210 content_html
= more
['content_html']
1211 more_widget_html
= more
['load_more_widget_html']
1213 playlist_title
= self
._html
_search
_regex
(
1214 r
'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1217 url_results
= self
._ids
_to
_results
(ids
)
1218 return self
.playlist_result(url_results
, playlist_id
, playlist_title
)
1221 class YoutubeTopListIE(YoutubePlaylistIE
):
1222 IE_NAME
= 'youtube:toplist'
1223 IE_DESC
= ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1224 ' (Example: "yttoplist:music:Top Tracks")')
1225 _VALID_URL
= r
'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1227 'url': 'yttoplist:music:Trending',
1228 'playlist_mincount': 5,
1229 'skip': 'Only works for logged-in users',
1232 def _real_extract(self
, url
):
1233 mobj
= re
.match(self
._VALID
_URL
, url
)
1234 channel
= mobj
.group('chann')
1235 title
= mobj
.group('title')
1236 query
= compat_urllib_parse
.urlencode({'title': title}
)
1237 channel_page
= self
._download
_webpage
(
1238 'https://www.youtube.com/%s' % channel
, title
)
1239 link
= self
._html
_search
_regex
(
1241 <a\s+href="([^"]+)".*?>\s*
1242 <span\s+class="branded-page-module-title-text">\s*
1243 <span[^>]*>.*?%s.*?</span>''' % re
.escape(query
),
1244 channel_page
, 'list')
1245 url
= compat_urlparse
.urljoin('https://www.youtube.com/', link
)
1247 video_re
= r
'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1249 # sometimes the webpage doesn't contain the videos
1250 # retry until we get them
1251 for i
in itertools
.count(0):
1252 msg
= 'Downloading Youtube mix'
1254 msg
+= ', retry #%d' % i
1256 webpage
= self
._download
_webpage
(url
, title
, msg
)
1257 ids
= orderedSet(re
.findall(video_re
, webpage
))
1260 url_results
= self
._ids
_to
_results
(ids
)
1261 return self
.playlist_result(url_results
, playlist_title
=title
)
1264 class YoutubeChannelIE(InfoExtractor
):
1265 IE_DESC
= 'YouTube.com channels'
1266 _VALID_URL
= r
'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1267 _MORE_PAGES_INDICATOR
= 'yt-uix-load-more'
1268 _MORE_PAGES_URL
= 'https://www.youtube.com/c4_browse_ajax?action_load_more_videos=1&flow=list&paging=%s&view=0&sort=da&channel_id=%s'
1269 IE_NAME
= 'youtube:channel'
1271 'note': 'paginated channel',
1272 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1273 'playlist_mincount': 91,
1276 def extract_videos_from_page(self
, page
):
1278 for mobj
in re
.finditer(r
'href="/watch\?v=([0-9A-Za-z_-]+)&?', page
):
1279 if mobj
.group(1) not in ids_in_page
:
1280 ids_in_page
.append(mobj
.group(1))
1283 def _real_extract(self
, url
):
1284 channel_id
= self
._match
_id
(url
)
1287 url
= 'https://www.youtube.com/channel/%s/videos' % channel_id
1288 channel_page
= self
._download
_webpage
(url
, channel_id
)
1289 autogenerated
= re
.search(r
'''(?x)
1291 channel-header-autogenerated-label|
1292 yt-channel-title-autogenerated
1293 )[^"]*"''', channel_page
) is not None
1296 # The videos are contained in a single page
1297 # the ajax pages can't be used, they are empty
1298 video_ids
= self
.extract_videos_from_page(channel_page
)
1300 self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1301 for video_id
in video_ids
]
1302 return self
.playlist_result(entries
, channel_id
)
1305 for pagenum
in itertools
.count(1):
1306 url
= self
._MORE
_PAGES
_URL
% (pagenum
, channel_id
)
1307 page
= self
._download
_json
(
1308 url
, channel_id
, note
='Downloading page #%s' % pagenum
,
1309 transform_source
=uppercase_escape
)
1311 ids_in_page
= self
.extract_videos_from_page(page
['content_html'])
1312 for video_id
in ids_in_page
:
1313 yield self
.url_result(
1314 video_id
, 'Youtube', video_id
=video_id
)
1316 if self
._MORE
_PAGES
_INDICATOR
not in page
['load_more_widget_html']:
1319 return self
.playlist_result(_entries(), channel_id
)
1322 class YoutubeUserIE(InfoExtractor
):
1323 IE_DESC
= 'YouTube.com user videos (URL or "ytuser" keyword)'
1324 _VALID_URL
= r
'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
1325 _TEMPLATE_URL
= 'https://gdata.youtube.com/feeds/api/users/%s'
1326 _GDATA_PAGE_SIZE
= 50
1327 _GDATA_URL
= 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1328 IE_NAME
= 'youtube:user'
1331 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1332 'playlist_mincount': 320,
1334 'title': 'TheLinuxFoundation',
1337 'url': 'ytuser:phihag',
1338 'only_matching': True,
1342 def suitable(cls
, url
):
1343 # Don't return True if the url can be extracted with other youtube
1344 # extractor, the regex would is too permissive and it would match.
1345 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1346 if any(ie
.suitable(url
) for ie
in other_ies
):
1349 return super(YoutubeUserIE
, cls
).suitable(url
)
1351 def _real_extract(self
, url
):
1352 username
= self
._match
_id
(url
)
1354 # Download video ids using YouTube Data API. Result size per
1355 # query is limited (currently to 50 videos) so we need to query
1356 # page by page until there are no video ids - it means we got
1359 def download_page(pagenum
):
1360 start_index
= pagenum
* self
._GDATA
_PAGE
_SIZE
+ 1
1362 gdata_url
= self
._GDATA
_URL
% (username
, self
._GDATA
_PAGE
_SIZE
, start_index
)
1363 page
= self
._download
_webpage
(
1364 gdata_url
, username
,
1365 'Downloading video ids from %d to %d' % (
1366 start_index
, start_index
+ self
._GDATA
_PAGE
_SIZE
))
1369 response
= json
.loads(page
)
1370 except ValueError as err
:
1371 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err
))
1372 if 'entry' not in response
['feed']:
1375 # Extract video identifiers
1376 entries
= response
['feed']['entry']
1377 for entry
in entries
:
1378 title
= entry
['title']['$t']
1379 video_id
= entry
['id']['$t'].split('/')[-1]
1383 'ie_key': 'Youtube',
1387 url_results
= OnDemandPagedList(download_page
, self
._GDATA
_PAGE
_SIZE
)
1389 return self
.playlist_result(url_results
, playlist_title
=username
)
1392 class YoutubeSearchIE(SearchInfoExtractor
):
1393 IE_DESC
= 'YouTube.com searches'
1394 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1396 IE_NAME
= 'youtube:search'
1397 _SEARCH_KEY
= 'ytsearch'
1399 def _get_n_results(self
, query
, n
):
1400 """Get a specified number of results for a query"""
1407 while (PAGE_SIZE
* pagenum
) < limit
:
1408 result_url
= self
._API
_URL
% (
1409 compat_urllib_parse
.quote_plus(query
.encode('utf-8')),
1410 (PAGE_SIZE
* pagenum
) + 1)
1411 data_json
= self
._download
_webpage
(
1412 result_url
, video_id
='query "%s"' % query
,
1413 note
='Downloading page %s' % (pagenum
+ 1),
1414 errnote
='Unable to download API page')
1415 data
= json
.loads(data_json
)
1416 api_response
= data
['data']
1418 if 'items' not in api_response
:
1419 raise ExtractorError(
1420 '[youtube] No video results', expected
=True)
1422 new_ids
= list(video
['id'] for video
in api_response
['items'])
1423 video_ids
+= new_ids
1425 limit
= min(n
, api_response
['totalItems'])
1428 if len(video_ids
) > n
:
1429 video_ids
= video_ids
[:n
]
1430 videos
= [self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1431 for video_id
in video_ids
]
1432 return self
.playlist_result(videos
, query
)
1435 class YoutubeSearchDateIE(YoutubeSearchIE
):
1436 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1437 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1438 _SEARCH_KEY
= 'ytsearchdate'
1439 IE_DESC
= 'YouTube.com searches, newest videos first'
1442 class YoutubeSearchURLIE(InfoExtractor
):
1443 IE_DESC
= 'YouTube.com search URLs'
1444 IE_NAME
= 'youtube:search_url'
1445 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1447 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1448 'playlist_mincount': 5,
1450 'title': 'youtube-dl test video',
1454 def _real_extract(self
, url
):
1455 mobj
= re
.match(self
._VALID
_URL
, url
)
1456 query
= compat_urllib_parse
.unquote_plus(mobj
.group('query'))
1458 webpage
= self
._download
_webpage
(url
, query
)
1459 result_code
= self
._search
_regex
(
1460 r
'(?s)<ol class="item-section"(.*?)</ol>', webpage
, 'result HTML')
1462 part_codes
= re
.findall(
1463 r
'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code
)
1465 for part_code
in part_codes
:
1466 part_title
= self
._html
_search
_regex
(
1467 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1468 part_url_snippet
= self
._html
_search
_regex
(
1469 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1470 part_url
= compat_urlparse
.urljoin(
1471 'https://www.youtube.com/', part_url_snippet
)
1475 'title': part_title
,
1479 '_type': 'playlist',
1485 class YoutubeShowIE(InfoExtractor
):
1486 IE_DESC
= 'YouTube.com (multi-season) shows'
1487 _VALID_URL
= r
'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1488 IE_NAME
= 'youtube:show'
1490 'url': 'http://www.youtube.com/show/airdisasters',
1491 'playlist_mincount': 3,
1493 'id': 'airdisasters',
1494 'title': 'Air Disasters',
1498 def _real_extract(self
, url
):
1499 mobj
= re
.match(self
._VALID
_URL
, url
)
1500 playlist_id
= mobj
.group('id')
1501 webpage
= self
._download
_webpage
(
1502 url
, playlist_id
, 'Downloading show webpage')
1503 # There's one playlist for each season of the show
1504 m_seasons
= list(re
.finditer(r
'href="(/playlist\?list=.*?)"', webpage
))
1505 self
.to_screen('%s: Found %s seasons' % (playlist_id
, len(m_seasons
)))
1508 'https://www.youtube.com' + season
.group(1), 'YoutubePlaylist')
1509 for season
in m_seasons
1511 title
= self
._og
_search
_title
(webpage
, fatal
=False)
1514 '_type': 'playlist',
1521 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1523 Base class for extractors that fetch info from
1524 http://www.youtube.com/feed_ajax
1525 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1527 _LOGIN_REQUIRED
= True
1528 # use action_load_personal_feed instead of action_load_system_feed
1529 _PERSONAL_FEED
= False
1532 def _FEED_TEMPLATE(self
):
1533 action
= 'action_load_system_feed'
1534 if self
._PERSONAL
_FEED
:
1535 action
= 'action_load_personal_feed'
1536 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action
, self
._FEED
_NAME
)
1540 return 'youtube:%s' % self
._FEED
_NAME
1542 def _real_initialize(self
):
1545 def _real_extract(self
, url
):
1548 for i
in itertools
.count(1):
1549 info
= self
._download
_json
(self
._FEED
_TEMPLATE
% paging
,
1550 '%s feed' % self
._FEED
_NAME
,
1551 'Downloading page %s' % i
)
1552 feed_html
= info
.get('feed_html') or info
.get('content_html')
1553 load_more_widget_html
= info
.get('load_more_widget_html') or feed_html
1554 m_ids
= re
.finditer(r
'"/watch\?v=(.*?)["&]', feed_html
)
1555 ids
= orderedSet(m
.group(1) for m
in m_ids
)
1556 feed_entries
.extend(
1557 self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1558 for video_id
in ids
)
1560 r
'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1561 load_more_widget_html
)
1564 paging
= mobj
.group('paging')
1565 return self
.playlist_result(feed_entries
, playlist_title
=self
._PLAYLIST
_TITLE
)
1568 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
1569 IE_DESC
= 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1570 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1571 _FEED_NAME
= 'recommended'
1572 _PLAYLIST_TITLE
= 'Youtube Recommended videos'
1575 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor
):
1576 IE_DESC
= 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1577 _VALID_URL
= r
'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1578 _FEED_NAME
= 'watch_later'
1579 _PLAYLIST_TITLE
= 'Youtube Watch Later'
1580 _PERSONAL_FEED
= True
1583 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
1584 IE_DESC
= 'Youtube watch history, ":ythistory" for short (requires authentication)'
1585 _VALID_URL
= 'https?://www\.youtube\.com/feed/history|:ythistory'
1586 _FEED_NAME
= 'history'
1587 _PERSONAL_FEED
= True
1588 _PLAYLIST_TITLE
= 'Youtube Watch History'
1591 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
1592 IE_NAME
= 'youtube:favorites'
1593 IE_DESC
= 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1594 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1595 _LOGIN_REQUIRED
= True
1597 def _real_extract(self
, url
):
1598 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1599 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, 'favourites playlist id')
1600 return self
.url_result(playlist_id
, 'YoutubePlaylist')
1603 class YoutubeSubscriptionsIE(YoutubePlaylistIE
):
1604 IE_NAME
= 'youtube:subscriptions'
1605 IE_DESC
= 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1606 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1609 def _real_extract(self
, url
):
1610 title
= 'Youtube Subscriptions'
1611 page
= self
._download
_webpage
('https://www.youtube.com/feed/subscriptions', title
)
1613 # The extraction process is the same as for playlists, but the regex
1614 # for the video ids doesn't contain an index
1616 more_widget_html
= content_html
= page
1618 for page_num
in itertools
.count(1):
1619 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
1620 new_ids
= orderedSet(matches
)
1623 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1627 more
= self
._download
_json
(
1628 'https://youtube.com/%s' % mobj
.group('more'), title
,
1629 'Downloading page #%s' % page_num
,
1630 transform_source
=uppercase_escape
)
1631 content_html
= more
['content_html']
1632 more_widget_html
= more
['load_more_widget_html']
1635 '_type': 'playlist',
1637 'entries': self
._ids
_to
_results
(ids
),
1641 class YoutubeTruncatedURLIE(InfoExtractor
):
1642 IE_NAME
= 'youtube:truncated_url'
1643 IE_DESC
= False # Do not list
1644 _VALID_URL
= r
'''(?x)
1645 (?:https?://)?[^/]+/watch\?(?:
1647 annotation_id=annotation_[^&]+
1649 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1653 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1654 'only_matching': True,
1656 'url': 'http://www.youtube.com/watch?',
1657 'only_matching': True,
1660 def _real_extract(self
, url
):
1661 raise ExtractorError(
1662 'Did you forget to quote the URL? Remember that & is a meta '
1663 'character in most shells, so you want to put the URL in quotes, '
1665 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1666 ' or simply youtube-dl BaW_jenozKc .',