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', 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
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', 'acodec': 'aac', 'vcodec': 'none', 'abr': 48, 'preference': -50},
268 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'vcodec': 'none', 'abr': 128, 'preference': -50},
269 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', '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 '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
291 '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'VP9'},
292 '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
295 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
296 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
298 # Dash webm audio with opus inside
299 '249': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50, 'preference': -50},
300 '250': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70, 'preference': -50},
301 '251': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160, 'preference': -50},
304 '_rtmp
': {'protocol': 'rtmp'},
310 'url
': 'http
://www
.youtube
.com
/watch?v
=BaW_jenozKc
',
314 'title
': 'youtube
-dl test video
"\'/\\ä↭𝕐',
315 'uploader': 'Philipp Hagemeister',
316 'uploader_id': 'phihag',
317 'upload_date': '20121002',
318 '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 .',
319 'categories
': ['Science
& Technology
'],
321 'dislike_count
': int,
325 'url
': 'http
://www
.youtube
.com
/watch?v
=UxxajLWwzqY
',
326 'note
': 'Test generic use_cipher_signature
video (#897)',
330 'upload_date': '20120506',
331 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
332 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
333 'uploader': 'Icona Pop',
334 'uploader_id': 'IconaPop',
338 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
339 'note': 'Test VEVO video with age protection (#956)',
343 'upload_date': '20130703',
344 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
345 'description': 'md5:64249768eec3bc4276236606ea996373',
346 'uploader': 'justintimberlakeVEVO',
347 'uploader_id': 'justintimberlakeVEVO',
351 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
352 'note': 'Embed-only video (#1746)',
356 'upload_date': '20120608',
357 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
358 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
359 'uploader': 'SET India',
360 'uploader_id': 'setindia'
364 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
365 'note': '256k DASH audio (format 141) via DASH manifest',
369 'upload_date': '20121002',
370 'uploader_id': '8KVIDEO',
372 'uploader': '8KVIDEO',
373 'title': 'UHDTV TEST 8K VIDEO.mp4'
376 'youtube_include_dash_manifest': True,
380 # DASH manifest with encrypted signature
382 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
386 'title': 'Afrojack, Spree Wilson - The Spark ft. Spree Wilson',
387 'description': 'md5:12e7067fa6735a77bdcbb58cb1187d2d',
388 'uploader': 'AfrojackVEVO',
389 'uploader_id': 'AfrojackVEVO',
390 'upload_date': '20131011',
393 'youtube_include_dash_manifest': True,
399 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
403 'upload_date': '20100909',
404 'uploader': 'The Amazing Atheist',
405 'uploader_id': 'TheAmazingAtheist',
406 'title': 'Burning Everyone\'s Koran',
407 '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',
410 # Normal age-gate video (No vevo, embed allowed)
412 'url': 'http://youtube.com/watch?v=HtVdAasjOgU',
416 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
417 'description': 're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
418 'uploader': 'The Witcher',
419 'uploader_id': 'WitcherGame',
420 'upload_date': '20140605',
423 # Age-gate video with encrypted signature
425 'url': 'http://www.youtube.com/watch?v=6kLq3WMV1nU',
429 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
430 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
431 'uploader': 'LloydVEVO',
432 'uploader_id': 'LloydVEVO',
433 'upload_date': '20110629',
436 # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
438 'url': '__2ABJjxzNo',
442 'upload_date': '20100430',
443 'uploader_id': 'deadmau5',
444 'description': 'md5:12c56784b8032162bb936a5f76d55360',
445 'uploader': 'deadmau5',
446 'title': 'Deadmau5 - Some Chords (HD)',
448 'expected_warnings': [
449 'DASH manifest missing',
452 # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
454 'url': 'lqQg6PlCWgI',
458 'upload_date': '20120731',
459 'uploader_id': 'olympic',
460 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
461 'uploader': 'Olympics',
462 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games',
465 'skip_download': 'requires avconv',
470 def __init__(self
, *args
, **kwargs
):
471 super(YoutubeIE
, self
).__init
__(*args
, **kwargs
)
472 self
._player
_cache
= {}
474 def report_video_info_webpage_download(self
, video_id
):
475 """Report attempt to download video info webpage."""
476 self
.to_screen('%s: Downloading video info webpage' % video_id
)
478 def report_information_extraction(self
, video_id
):
479 """Report attempt to extract video information."""
480 self
.to_screen('%s: Extracting video information' % video_id
)
482 def report_unavailable_format(self
, video_id
, format
):
483 """Report extracted video URL."""
484 self
.to_screen('%s: Format %s not available' % (video_id
, format
))
486 def report_rtmp_download(self
):
487 """Indicate the download will use the RTMP protocol."""
488 self
.to_screen('RTMP download detected')
490 def _signature_cache_id(self
, example_sig
):
491 """ Return a string representation of a signature """
492 return '.'.join(compat_str(len(part
)) for part
in example_sig
.split('.'))
494 def _extract_signature_function(self
, video_id
, player_url
, example_sig
):
496 r
'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
499 raise ExtractorError('Cannot identify player %r' % player_url
)
500 player_type
= id_m
.group('ext')
501 player_id
= id_m
.group('id')
503 # Read from filesystem cache
504 func_id
= '%s_%s_%s' % (
505 player_type
, player_id
, self
._signature
_cache
_id
(example_sig
))
506 assert os
.path
.basename(func_id
) == func_id
508 cache_spec
= self
._downloader
.cache
.load('youtube-sigfuncs', func_id
)
509 if cache_spec
is not None:
510 return lambda s
: ''.join(s
[i
] for i
in cache_spec
)
512 if player_type
== 'js':
513 code
= self
._download
_webpage
(
514 player_url
, video_id
,
515 note
='Downloading %s player %s' % (player_type
, player_id
),
516 errnote
='Download of %s failed' % player_url
)
517 res
= self
._parse
_sig
_js
(code
)
518 elif player_type
== 'swf':
519 urlh
= self
._request
_webpage
(
520 player_url
, video_id
,
521 note
='Downloading %s player %s' % (player_type
, player_id
),
522 errnote
='Download of %s failed' % player_url
)
524 res
= self
._parse
_sig
_swf
(code
)
526 assert False, 'Invalid player type %r' % player_type
528 if cache_spec
is None:
529 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
530 cache_res
= res(test_string
)
531 cache_spec
= [ord(c
) for c
in cache_res
]
533 self
._downloader
.cache
.store('youtube-sigfuncs', func_id
, cache_spec
)
536 def _print_sig_code(self
, func
, example_sig
):
537 def gen_sig_code(idxs
):
538 def _genslice(start
, end
, step
):
539 starts
= '' if start
== 0 else str(start
)
540 ends
= (':%d' % (end
+ step
)) if end
+ step
>= 0 else ':'
541 steps
= '' if step
== 1 else (':%d' % step
)
542 return 's[%s%s%s]' % (starts
, ends
, steps
)
545 # Quelch pyflakes warnings - start will be set when step is set
546 start
= '(Never used)'
547 for i
, prev
in zip(idxs
[1:], idxs
[:-1]):
551 yield _genslice(start
, prev
, step
)
554 if i
- prev
in [-1, 1]:
563 yield _genslice(start
, i
, step
)
565 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
566 cache_res
= func(test_string
)
567 cache_spec
= [ord(c
) for c
in cache_res
]
568 expr_code
= ' + '.join(gen_sig_code(cache_spec
))
569 signature_id_tuple
= '(%s)' % (
570 ', '.join(compat_str(len(p
)) for p
in example_sig
.split('.')))
571 code
= ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
572 ' return %s\n') % (signature_id_tuple
, expr_code
)
573 self
.to_screen('Extracted signature function:\n' + code
)
575 def _parse_sig_js(self
, jscode
):
576 funcname
= self
._search
_regex
(
577 r
'\.sig\|\|([a-zA-Z0-9]+)\(', jscode
,
578 'Initial JS player signature function name')
580 jsi
= JSInterpreter(jscode
)
581 initial_function
= jsi
.extract_function(funcname
)
582 return lambda s
: initial_function([s
])
584 def _parse_sig_swf(self
, file_contents
):
585 swfi
= SWFInterpreter(file_contents
)
586 TARGET_CLASSNAME
= 'SignatureDecipher'
587 searched_class
= swfi
.extract_class(TARGET_CLASSNAME
)
588 initial_function
= swfi
.extract_function(searched_class
, 'decipher')
589 return lambda s
: initial_function([s
])
591 def _decrypt_signature(self
, s
, video_id
, player_url
, age_gate
=False):
592 """Turn the encrypted s field into a working signature"""
594 if player_url
is None:
595 raise ExtractorError('Cannot decrypt signature without player_url')
597 if player_url
.startswith('//'):
598 player_url
= 'https:' + player_url
600 player_id
= (player_url
, self
._signature
_cache
_id
(s
))
601 if player_id
not in self
._player
_cache
:
602 func
= self
._extract
_signature
_function
(
603 video_id
, player_url
, s
605 self
._player
_cache
[player_id
] = func
606 func
= self
._player
_cache
[player_id
]
607 if self
._downloader
.params
.get('youtube_print_sig_code'):
608 self
._print
_sig
_code
(func
, s
)
610 except Exception as e
:
611 tb
= traceback
.format_exc()
612 raise ExtractorError(
613 'Signature extraction failed: ' + tb
, cause
=e
)
615 def _get_available_subtitles(self
, video_id
, webpage
):
617 subs_doc
= self
._download
_xml
(
618 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
,
619 video_id
, note
=False)
620 except ExtractorError
as err
:
621 self
._downloader
.report_warning('unable to download video subtitles: %s' % compat_str(err
))
625 for track
in subs_doc
.findall('track'):
626 lang
= track
.attrib
['lang_code']
627 if lang
in sub_lang_list
:
629 params
= compat_urllib_parse
.urlencode({
632 'fmt': self
._downloader
.params
.get('subtitlesformat', 'srt'),
633 'name': track
.attrib
['name'].encode('utf-8'),
635 url
= 'https://www.youtube.com/api/timedtext?' + params
636 sub_lang_list
[lang
] = url
637 if not sub_lang_list
:
638 self
._downloader
.report_warning('video doesn\'t have subtitles')
642 def _get_available_automatic_caption(self
, video_id
, webpage
):
643 """We need the webpage for getting the captions url, pass it as an
644 argument to speed up the process."""
645 sub_format
= self
._downloader
.params
.get('subtitlesformat', 'srt')
646 self
.to_screen('%s: Looking for automatic captions' % video_id
)
647 mobj
= re
.search(r
';ytplayer.config = ({.*?});', webpage
)
648 err_msg
= 'Couldn\'t find automatic captions for %s' % video_id
650 self
._downloader
.report_warning(err_msg
)
652 player_config
= json
.loads(mobj
.group(1))
654 args
= player_config
['args']
655 caption_url
= args
['ttsurl']
656 timestamp
= args
['timestamp']
657 # We get the available subtitles
658 list_params
= compat_urllib_parse
.urlencode({
663 list_url
= caption_url
+ '&' + list_params
664 caption_list
= self
._download
_xml
(list_url
, video_id
)
665 original_lang_node
= caption_list
.find('track')
666 if original_lang_node
is None:
667 self
._downloader
.report_warning('Video doesn\'t have automatic captions')
669 original_lang
= original_lang_node
.attrib
['lang_code']
670 caption_kind
= original_lang_node
.attrib
.get('kind', '')
673 for lang_node
in caption_list
.findall('target'):
674 sub_lang
= lang_node
.attrib
['lang_code']
675 params
= compat_urllib_parse
.urlencode({
676 'lang': original_lang
,
680 'kind': caption_kind
,
682 sub_lang_list
[sub_lang
] = caption_url
+ '&' + params
684 # An extractor error can be raise by the download process if there are
685 # no automatic captions but there are subtitles
686 except (KeyError, ExtractorError
):
687 self
._downloader
.report_warning(err_msg
)
691 def extract_id(cls
, url
):
692 mobj
= re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
)
694 raise ExtractorError('Invalid URL: %s' % url
)
695 video_id
= mobj
.group(2)
698 def _extract_from_m3u8(self
, manifest_url
, video_id
):
701 def _get_urls(_manifest
):
702 lines
= _manifest
.split('\n')
703 urls
= filter(lambda l
: l
and not l
.startswith('#'),
706 manifest
= self
._download
_webpage
(manifest_url
, video_id
, 'Downloading formats manifest')
707 formats_urls
= _get_urls(manifest
)
708 for format_url
in formats_urls
:
709 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
710 url_map
[itag
] = format_url
713 def _extract_annotations(self
, video_id
):
714 url
= 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
715 return self
._download
_webpage
(url
, video_id
, note
='Searching for annotations.', errnote
='Unable to download video annotations.')
717 def _parse_dash_manifest(
718 self
, video_id
, dash_manifest_url
, player_url
, age_gate
):
719 def decrypt_sig(mobj
):
721 dec_s
= self
._decrypt
_signature
(s
, video_id
, player_url
, age_gate
)
722 return '/signature/%s' % dec_s
723 dash_manifest_url
= re
.sub(r
'/s/([\w\.]+)', decrypt_sig
, dash_manifest_url
)
724 dash_doc
= self
._download
_xml
(
725 dash_manifest_url
, video_id
,
726 note
='Downloading DASH manifest',
727 errnote
='Could not download DASH manifest')
730 for r
in dash_doc
.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
731 url_el
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
734 format_id
= r
.attrib
['id']
735 video_url
= url_el
.text
736 filesize
= int_or_none(url_el
.attrib
.get('{http://youtube.com/yt/2012/10/10}contentLength'))
738 'format_id': format_id
,
740 'width': int_or_none(r
.attrib
.get('width')),
741 'height': int_or_none(r
.attrib
.get('height')),
742 'tbr': int_or_none(r
.attrib
.get('bandwidth'), 1000),
743 'asr': int_or_none(r
.attrib
.get('audioSamplingRate')),
744 'filesize': filesize
,
745 'fps': int_or_none(r
.attrib
.get('frameRate')),
748 existing_format
= next(
750 if fo
['format_id'] == format_id
)
751 except StopIteration:
752 f
.update(self
._formats
.get(format_id
, {}).items())
755 existing_format
.update(f
)
758 def _real_extract(self
, url
):
760 'http' if self
._downloader
.params
.get('prefer_insecure', False)
763 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
764 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
766 url
= proto
+ '://www.youtube.com/' + compat_urllib_parse
.unquote(mobj
.group(1)).lstrip('/')
767 video_id
= self
.extract_id(url
)
770 url
= proto
+ '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
771 video_webpage
= self
._download
_webpage
(url
, video_id
)
773 # Attempt to extract SWF player URL
774 mobj
= re
.search(r
'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
776 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
781 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
783 # We simulate the access to the video from www.youtube.com/v/{video_id}
784 # this can be viewed without login into Youtube
785 url
= proto
+ '://www.youtube.com/embed/%s' % video_id
786 embed_webpage
= self
._download
_webpage
(url
, video_id
, 'Downloading embed webpage')
787 data
= compat_urllib_parse
.urlencode({
788 'video_id': video_id
,
789 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
790 'sts': self
._search
_regex
(
791 r
'"sts"\s*:\s*(\d+)', embed_webpage
, 'sts', default
=''),
793 video_info_url
= proto
+ '://www.youtube.com/get_video_info?' + data
794 video_info_webpage
= self
._download
_webpage
(
795 video_info_url
, video_id
,
796 note
='Refetching age-gated info webpage',
797 errnote
='unable to download video info webpage')
798 video_info
= compat_parse_qs(video_info_webpage
)
802 # Try looking directly into the video webpage
803 mobj
= re
.search(r
';ytplayer\.config\s*=\s*({.*?});', video_webpage
)
805 raise ValueError('Could not find ytplayer.config') # caught below
806 json_code
= uppercase_escape(mobj
.group(1))
807 ytplayer_config
= json
.loads(json_code
)
808 args
= ytplayer_config
['args']
809 # Convert to the same format returned by compat_parse_qs
810 video_info
= dict((k
, [v
]) for k
, v
in args
.items())
811 if 'url_encoded_fmt_stream_map' not in args
:
812 raise ValueError('No stream_map present') # caught below
814 # We fallback to the get_video_info pages (used by the embed page)
815 self
.report_video_info_webpage_download(video_id
)
816 for el_type
in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
818 '%s://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
819 % (proto
, video_id
, el_type
))
820 video_info_webpage
= self
._download
_webpage
(
822 video_id
, note
=False,
823 errnote
='unable to download video info webpage')
824 video_info
= compat_parse_qs(video_info_webpage
)
825 if 'token' in video_info
:
827 if 'token' not in video_info
:
828 if 'reason' in video_info
:
829 raise ExtractorError(
830 'YouTube said: %s' % video_info
['reason'][0],
831 expected
=True, video_id
=video_id
)
833 raise ExtractorError(
834 '"token" parameter not in video info for unknown reason',
837 if 'view_count' in video_info
:
838 view_count
= int(video_info
['view_count'][0])
842 # Check for "rental" videos
843 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
844 raise ExtractorError('"rental" videos not supported')
846 # Start extracting information
847 self
.report_information_extraction(video_id
)
850 if 'author' not in video_info
:
851 raise ExtractorError('Unable to extract uploader name')
852 video_uploader
= compat_urllib_parse
.unquote_plus(video_info
['author'][0])
855 video_uploader_id
= None
856 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
858 video_uploader_id
= mobj
.group(1)
860 self
._downloader
.report_warning('unable to extract uploader nickname')
863 if 'title' in video_info
:
864 video_title
= video_info
['title'][0]
866 self
._downloader
.report_warning('Unable to extract video title')
870 # We try first to get a high quality image:
871 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
872 video_webpage
, re
.DOTALL
)
873 if m_thumb
is not None:
874 video_thumbnail
= m_thumb
.group(1)
875 elif 'thumbnail_url' not in video_info
:
876 self
._downloader
.report_warning('unable to extract video thumbnail')
877 video_thumbnail
= None
878 else: # don't panic if we can't find it
879 video_thumbnail
= compat_urllib_parse
.unquote_plus(video_info
['thumbnail_url'][0])
883 mobj
= re
.search(r
'(?s)id="eow-date.*?>(.*?)</span>', video_webpage
)
886 r
'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
889 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
890 upload_date
= unified_strdate(upload_date
)
892 m_cat_container
= self
._search
_regex
(
893 r
'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
894 video_webpage
, 'categories', default
=None)
896 category
= self
._html
_search
_regex
(
897 r
'(?s)<a[^<]+>(.*?)</a>', m_cat_container
, 'category',
899 video_categories
= None if category
is None else [category
]
901 video_categories
= None
904 video_description
= get_element_by_id("eow-description", video_webpage
)
905 if video_description
:
906 video_description
= re
.sub(r
'''(?x)
908 (?:[a-zA-Z-]+="[^"]+"\s+)*?
910 (?:[a-zA-Z-]+="[^"]+"\s+)*?
911 class="yt-uix-redirect-link"\s*>
914 ''', r
'\1', video_description
)
915 video_description
= clean_html(video_description
)
917 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
919 video_description
= unescapeHTML(fd_mobj
.group(1))
921 video_description
= ''
923 def _extract_count(count_name
):
924 count
= self
._search
_regex
(
925 r
'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re
.escape(count_name
),
926 video_webpage
, count_name
, default
=None)
927 if count
is not None:
928 return int(count
.replace(',', ''))
930 like_count
= _extract_count('like')
931 dislike_count
= _extract_count('dislike')
934 video_subtitles
= self
.extract_subtitles(video_id
, video_webpage
)
936 if self
._downloader
.params
.get('listsubtitles', False):
937 self
._list
_available
_subtitles
(video_id
, video_webpage
)
940 if 'length_seconds' not in video_info
:
941 self
._downloader
.report_warning('unable to extract video duration')
942 video_duration
= None
944 video_duration
= int(compat_urllib_parse
.unquote_plus(video_info
['length_seconds'][0]))
947 video_annotations
= None
948 if self
._downloader
.params
.get('writeannotations', False):
949 video_annotations
= self
._extract
_annotations
(video_id
)
951 def _map_to_format_list(urlmap
):
953 for itag
, video_real_url
in urlmap
.items():
956 'url': video_real_url
,
957 'player_url': player_url
,
959 if itag
in self
._formats
:
960 dct
.update(self
._formats
[itag
])
964 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
965 self
.report_rtmp_download()
967 'format_id': '_rtmp',
969 'url': video_info
['conn'][0],
970 'player_url': player_url
,
972 elif len(video_info
.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info
.get('adaptive_fmts', [''])[0]) >= 1:
973 encoded_url_map
= video_info
.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info
.get('adaptive_fmts', [''])[0]
974 if 'rtmpe%3Dyes' in encoded_url_map
:
975 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
977 for url_data_str
in encoded_url_map
.split(','):
978 url_data
= compat_parse_qs(url_data_str
)
979 if 'itag' not in url_data
or 'url' not in url_data
:
981 format_id
= url_data
['itag'][0]
982 url
= url_data
['url'][0]
984 if 'sig' in url_data
:
985 url
+= '&signature=' + url_data
['sig'][0]
986 elif 's' in url_data
:
987 encrypted_sig
= url_data
['s'][0]
989 jsplayer_url_json
= self
._search
_regex
(
990 r
'"assets":.+?"js":\s*("[^"]+")',
991 embed_webpage
if age_gate
else video_webpage
, 'JS player URL')
992 player_url
= json
.loads(jsplayer_url_json
)
993 if player_url
is None:
994 player_url_json
= self
._search
_regex
(
995 r
'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
996 video_webpage
, 'age gate player URL')
997 player_url
= json
.loads(player_url_json
)
999 if self
._downloader
.params
.get('verbose'):
1000 if player_url
is None:
1001 player_version
= 'unknown'
1002 player_desc
= 'unknown'
1004 if player_url
.endswith('swf'):
1005 player_version
= self
._search
_regex
(
1006 r
'-(.+?)(?:/watch_as3)?\.swf$', player_url
,
1007 'flash player', fatal
=False)
1008 player_desc
= 'flash player %s' % player_version
1010 player_version
= self
._search
_regex
(
1011 r
'html5player-([^/]+?)(?:/html5player)?\.js',
1013 'html5 player', fatal
=False)
1014 player_desc
= 'html5 player %s' % player_version
1016 parts_sizes
= self
._signature
_cache
_id
(encrypted_sig
)
1017 self
.to_screen('{%s} signature length %s, %s' %
1018 (format_id
, parts_sizes
, player_desc
))
1020 signature
= self
._decrypt
_signature
(
1021 encrypted_sig
, video_id
, player_url
, age_gate
)
1022 url
+= '&signature=' + signature
1023 if 'ratebypass' not in url
:
1024 url
+= '&ratebypass=yes'
1025 url_map
[format_id
] = url
1026 formats
= _map_to_format_list(url_map
)
1027 elif video_info
.get('hlsvp'):
1028 manifest_url
= video_info
['hlsvp'][0]
1029 url_map
= self
._extract
_from
_m
3u8(manifest_url
, video_id
)
1030 formats
= _map_to_format_list(url_map
)
1032 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
1034 # Look for the DASH manifest
1035 if self
._downloader
.params
.get('youtube_include_dash_manifest', True):
1036 dash_mpd
= video_info
.get('dashmpd')
1038 dash_manifest_url
= dash_mpd
[0]
1040 dash_formats
= self
._parse
_dash
_manifest
(
1041 video_id
, dash_manifest_url
, player_url
, age_gate
)
1042 except (ExtractorError
, KeyError) as e
:
1043 self
.report_warning(
1044 'Skipping DASH manifest: %r' % e
, video_id
)
1046 # Hide the formats we found through non-DASH
1047 dash_keys
= set(df
['format_id'] for df
in dash_formats
)
1049 if f
['format_id'] in dash_keys
:
1050 f
['format_id'] = 'nondash-%s' % f
['format_id']
1051 f
['preference'] = f
.get('preference', 0) - 10000
1052 formats
.extend(dash_formats
)
1054 self
._sort
_formats
(formats
)
1058 'uploader': video_uploader
,
1059 'uploader_id': video_uploader_id
,
1060 'upload_date': upload_date
,
1061 'title': video_title
,
1062 'thumbnail': video_thumbnail
,
1063 'description': video_description
,
1064 'categories': video_categories
,
1065 'subtitles': video_subtitles
,
1066 'duration': video_duration
,
1067 'age_limit': 18 if age_gate
else 0,
1068 'annotations': video_annotations
,
1069 'webpage_url': proto
+ '://www.youtube.com/watch?v=%s' % video_id
,
1070 'view_count': view_count
,
1071 'like_count': like_count
,
1072 'dislike_count': dislike_count
,
1077 class YoutubePlaylistIE(YoutubeBaseInfoExtractor
):
1078 IE_DESC
= 'YouTube.com playlists'
1079 _VALID_URL
= r
"""(?x)(?:
1084 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1085 \? (?:.*?&)*? (?:p|a|list)=
1089 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1090 # Top tracks, they can also include dots
1095 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1097 _TEMPLATE_URL
= 'https://www.youtube.com/playlist?list=%s'
1098 _VIDEO_RE
= r
'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1099 IE_NAME
= 'youtube:playlist'
1101 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1103 'title': 'ytdl test PL',
1104 'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1106 'playlist_count': 3,
1108 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1110 'title': 'YDL_Empty_List',
1112 'playlist_count': 0,
1114 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1115 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1117 'title': '29C3: Not my department',
1119 'playlist_count': 95,
1121 'note': 'issue #673',
1122 'url': 'PLBB231211A4F62143',
1124 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1126 'playlist_mincount': 26,
1128 'note': 'Large playlist',
1129 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1131 'title': 'Uploads from Cauchemar',
1133 'playlist_mincount': 799,
1135 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1137 'title': 'YDL_safe_search',
1139 'playlist_count': 2,
1142 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1143 'playlist_count': 4,
1148 'note': 'Embedded SWF player',
1149 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1150 'playlist_count': 4,
1155 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
1156 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
1158 'title': 'Uploads from Interstellar Movie',
1160 'playlist_mincout': 21,
1163 def _real_initialize(self
):
1166 def _ids_to_results(self
, ids
):
1168 self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
1171 def _extract_mix(self
, playlist_id
):
1172 # The mixes are generated from a a single video
1173 # the id of the playlist is just 'RD' + video_id
1174 url
= 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id
[-11:], playlist_id
)
1175 webpage
= self
._download
_webpage
(
1176 url
, playlist_id
, 'Downloading Youtube mix')
1177 search_title
= lambda class_name
: get_element_by_attribute('class', class_name
, webpage
)
1179 search_title('playlist-title') or
1180 search_title('title long-title') or
1181 search_title('title'))
1182 title
= clean_html(title_span
)
1183 ids
= orderedSet(re
.findall(
1184 r
'''(?xs)data-video-username=".*?".*?
1185 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re
.escape(playlist_id
),
1187 url_results
= self
._ids
_to
_results
(ids
)
1189 return self
.playlist_result(url_results
, playlist_id
, title
)
1191 def _real_extract(self
, url
):
1192 # Extract playlist id
1193 mobj
= re
.match(self
._VALID
_URL
, url
)
1195 raise ExtractorError('Invalid URL: %s' % url
)
1196 playlist_id
= mobj
.group(1) or mobj
.group(2)
1198 # Check if it's a video-specific URL
1199 query_dict
= compat_urlparse
.parse_qs(compat_urlparse
.urlparse(url
).query
)
1200 if 'v' in query_dict
:
1201 video_id
= query_dict
['v'][0]
1202 if self
._downloader
.params
.get('noplaylist'):
1203 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1204 return self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1206 self
.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id
, video_id
))
1208 if playlist_id
.startswith('RD'):
1209 # Mixes require a custom extraction process
1210 return self
._extract
_mix
(playlist_id
)
1212 url
= self
._TEMPLATE
_URL
% playlist_id
1213 page
= self
._download
_webpage
(url
, playlist_id
)
1214 more_widget_html
= content_html
= page
1216 # Check if the playlist exists or is private
1217 if re
.search(r
'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page
) is not None:
1218 raise ExtractorError(
1219 'The playlist doesn\'t exist or is private, use --username or '
1220 '--netrc to access it.',
1223 # Extract the video ids from the playlist pages
1226 for page_num
in itertools
.count(1):
1227 matches
= re
.finditer(self
._VIDEO
_RE
, content_html
)
1228 # We remove the duplicates and the link with index 0
1229 # (it's not the first video of the playlist)
1230 new_ids
= orderedSet(m
.group('id') for m
in matches
if m
.group('index') != '0')
1233 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1237 more
= self
._download
_json
(
1238 'https://youtube.com/%s' % mobj
.group('more'), playlist_id
,
1239 'Downloading page #%s' % page_num
,
1240 transform_source
=uppercase_escape
)
1241 content_html
= more
['content_html']
1242 if not content_html
.strip():
1243 # Some webpages show a "Load more" button but they don't
1246 more_widget_html
= more
['load_more_widget_html']
1248 playlist_title
= self
._html
_search
_regex
(
1249 r
'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1252 url_results
= self
._ids
_to
_results
(ids
)
1253 return self
.playlist_result(url_results
, playlist_id
, playlist_title
)
1256 class YoutubeChannelIE(InfoExtractor
):
1257 IE_DESC
= 'YouTube.com channels'
1258 _VALID_URL
= r
'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
1259 IE_NAME
= 'youtube:channel'
1261 'note': 'paginated channel',
1262 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1263 'playlist_mincount': 91,
1266 def extract_videos_from_page(self
, page
):
1268 for mobj
in re
.finditer(r
'href="/watch\?v=([0-9A-Za-z_-]+)&?', page
):
1269 if mobj
.group(1) not in ids_in_page
:
1270 ids_in_page
.append(mobj
.group(1))
1273 def _real_extract(self
, url
):
1274 channel_id
= self
._match
_id
(url
)
1277 url
= 'https://www.youtube.com/channel/%s/videos' % channel_id
1278 channel_page
= self
._download
_webpage
(url
, channel_id
)
1279 autogenerated
= re
.search(r
'''(?x)
1281 channel-header-autogenerated-label|
1282 yt-channel-title-autogenerated
1283 )[^"]*"''', channel_page
) is not None
1286 # The videos are contained in a single page
1287 # the ajax pages can't be used, they are empty
1288 video_ids
= self
.extract_videos_from_page(channel_page
)
1290 self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1291 for video_id
in video_ids
]
1292 return self
.playlist_result(entries
, channel_id
)
1295 more_widget_html
= content_html
= channel_page
1296 for pagenum
in itertools
.count(1):
1298 ids_in_page
= self
.extract_videos_from_page(content_html
)
1299 for video_id
in ids_in_page
:
1300 yield self
.url_result(
1301 video_id
, 'Youtube', video_id
=video_id
)
1304 r
'data-uix-load-more-href="/?(?P<more>[^"]+)"',
1309 more
= self
._download
_json
(
1310 'https://youtube.com/%s' % mobj
.group('more'), channel_id
,
1311 'Downloading page #%s' % (pagenum
+ 1),
1312 transform_source
=uppercase_escape
)
1313 content_html
= more
['content_html']
1314 more_widget_html
= more
['load_more_widget_html']
1316 return self
.playlist_result(_entries(), channel_id
)
1319 class YoutubeUserIE(InfoExtractor
):
1320 IE_DESC
= 'YouTube.com user videos (URL or "ytuser" keyword)'
1321 _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_-]+)'
1322 _TEMPLATE_URL
= 'https://gdata.youtube.com/feeds/api/users/%s'
1323 _GDATA_PAGE_SIZE
= 50
1324 _GDATA_URL
= 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1325 IE_NAME
= 'youtube:user'
1328 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1329 'playlist_mincount': 320,
1331 'title': 'TheLinuxFoundation',
1334 'url': 'ytuser:phihag',
1335 'only_matching': True,
1339 def suitable(cls
, url
):
1340 # Don't return True if the url can be extracted with other youtube
1341 # extractor, the regex would is too permissive and it would match.
1342 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1343 if any(ie
.suitable(url
) for ie
in other_ies
):
1346 return super(YoutubeUserIE
, cls
).suitable(url
)
1348 def _real_extract(self
, url
):
1349 username
= self
._match
_id
(url
)
1351 # Download video ids using YouTube Data API. Result size per
1352 # query is limited (currently to 50 videos) so we need to query
1353 # page by page until there are no video ids - it means we got
1356 def download_page(pagenum
):
1357 start_index
= pagenum
* self
._GDATA
_PAGE
_SIZE
+ 1
1359 gdata_url
= self
._GDATA
_URL
% (username
, self
._GDATA
_PAGE
_SIZE
, start_index
)
1360 page
= self
._download
_webpage
(
1361 gdata_url
, username
,
1362 'Downloading video ids from %d to %d' % (
1363 start_index
, start_index
+ self
._GDATA
_PAGE
_SIZE
))
1366 response
= json
.loads(page
)
1367 except ValueError as err
:
1368 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err
))
1369 if 'entry' not in response
['feed']:
1372 # Extract video identifiers
1373 entries
= response
['feed']['entry']
1374 for entry
in entries
:
1375 title
= entry
['title']['$t']
1376 video_id
= entry
['id']['$t'].split('/')[-1]
1380 'ie_key': 'Youtube',
1384 url_results
= OnDemandPagedList(download_page
, self
._GDATA
_PAGE
_SIZE
)
1386 return self
.playlist_result(url_results
, playlist_title
=username
)
1389 class YoutubeSearchIE(SearchInfoExtractor
):
1390 IE_DESC
= 'YouTube.com searches'
1391 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1393 IE_NAME
= 'youtube:search'
1394 _SEARCH_KEY
= 'ytsearch'
1396 def _get_n_results(self
, query
, n
):
1397 """Get a specified number of results for a query"""
1404 while (PAGE_SIZE
* pagenum
) < limit
:
1405 result_url
= self
._API
_URL
% (
1406 compat_urllib_parse
.quote_plus(query
.encode('utf-8')),
1407 (PAGE_SIZE
* pagenum
) + 1)
1408 data_json
= self
._download
_webpage
(
1409 result_url
, video_id
='query "%s"' % query
,
1410 note
='Downloading page %s' % (pagenum
+ 1),
1411 errnote
='Unable to download API page')
1412 data
= json
.loads(data_json
)
1413 api_response
= data
['data']
1415 if 'items' not in api_response
:
1416 raise ExtractorError(
1417 '[youtube] No video results', expected
=True)
1419 new_ids
= list(video
['id'] for video
in api_response
['items'])
1420 video_ids
+= new_ids
1422 limit
= min(n
, api_response
['totalItems'])
1425 if len(video_ids
) > n
:
1426 video_ids
= video_ids
[:n
]
1427 videos
= [self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1428 for video_id
in video_ids
]
1429 return self
.playlist_result(videos
, query
)
1432 class YoutubeSearchDateIE(YoutubeSearchIE
):
1433 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1434 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1435 _SEARCH_KEY
= 'ytsearchdate'
1436 IE_DESC
= 'YouTube.com searches, newest videos first'
1439 class YoutubeSearchURLIE(InfoExtractor
):
1440 IE_DESC
= 'YouTube.com search URLs'
1441 IE_NAME
= 'youtube:search_url'
1442 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1444 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1445 'playlist_mincount': 5,
1447 'title': 'youtube-dl test video',
1451 def _real_extract(self
, url
):
1452 mobj
= re
.match(self
._VALID
_URL
, url
)
1453 query
= compat_urllib_parse
.unquote_plus(mobj
.group('query'))
1455 webpage
= self
._download
_webpage
(url
, query
)
1456 result_code
= self
._search
_regex
(
1457 r
'(?s)<ol class="item-section"(.*?)</ol>', webpage
, 'result HTML')
1459 part_codes
= re
.findall(
1460 r
'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code
)
1462 for part_code
in part_codes
:
1463 part_title
= self
._html
_search
_regex
(
1464 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1465 part_url_snippet
= self
._html
_search
_regex
(
1466 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1467 part_url
= compat_urlparse
.urljoin(
1468 'https://www.youtube.com/', part_url_snippet
)
1472 'title': part_title
,
1476 '_type': 'playlist',
1482 class YoutubeShowIE(InfoExtractor
):
1483 IE_DESC
= 'YouTube.com (multi-season) shows'
1484 _VALID_URL
= r
'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1485 IE_NAME
= 'youtube:show'
1487 'url': 'http://www.youtube.com/show/airdisasters',
1488 'playlist_mincount': 3,
1490 'id': 'airdisasters',
1491 'title': 'Air Disasters',
1495 def _real_extract(self
, url
):
1496 mobj
= re
.match(self
._VALID
_URL
, url
)
1497 playlist_id
= mobj
.group('id')
1498 webpage
= self
._download
_webpage
(
1499 url
, playlist_id
, 'Downloading show webpage')
1500 # There's one playlist for each season of the show
1501 m_seasons
= list(re
.finditer(r
'href="(/playlist\?list=.*?)"', webpage
))
1502 self
.to_screen('%s: Found %s seasons' % (playlist_id
, len(m_seasons
)))
1505 'https://www.youtube.com' + season
.group(1), 'YoutubePlaylist')
1506 for season
in m_seasons
1508 title
= self
._og
_search
_title
(webpage
, fatal
=False)
1511 '_type': 'playlist',
1518 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1520 Base class for extractors that fetch info from
1521 http://www.youtube.com/feed_ajax
1522 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1524 _LOGIN_REQUIRED
= True
1525 # use action_load_personal_feed instead of action_load_system_feed
1526 _PERSONAL_FEED
= False
1529 def _FEED_TEMPLATE(self
):
1530 action
= 'action_load_system_feed'
1531 if self
._PERSONAL
_FEED
:
1532 action
= 'action_load_personal_feed'
1533 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action
, self
._FEED
_NAME
)
1537 return 'youtube:%s' % self
._FEED
_NAME
1539 def _real_initialize(self
):
1542 def _real_extract(self
, url
):
1545 for i
in itertools
.count(1):
1546 info
= self
._download
_json
(
1547 self
._FEED
_TEMPLATE
% paging
,
1548 '%s feed' % self
._FEED
_NAME
,
1549 'Downloading page %s' % i
,
1550 transform_source
=uppercase_escape
)
1551 feed_html
= info
.get('feed_html') or info
.get('content_html')
1552 load_more_widget_html
= info
.get('load_more_widget_html') or feed_html
1553 m_ids
= re
.finditer(r
'"/watch\?v=(.*?)["&]', feed_html
)
1554 ids
= orderedSet(m
.group(1) for m
in m_ids
)
1555 feed_entries
.extend(
1556 self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1557 for video_id
in ids
)
1559 r
'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1560 load_more_widget_html
)
1563 paging
= mobj
.group('paging')
1564 return self
.playlist_result(feed_entries
, playlist_title
=self
._PLAYLIST
_TITLE
)
1567 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
1568 IE_DESC
= 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
1569 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1570 _FEED_NAME
= 'recommended'
1571 _PLAYLIST_TITLE
= 'Youtube Recommended videos'
1574 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor
):
1575 IE_DESC
= 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
1576 _VALID_URL
= r
'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1577 _FEED_NAME
= 'watch_later'
1578 _PLAYLIST_TITLE
= 'Youtube Watch Later'
1579 _PERSONAL_FEED
= True
1582 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
1583 IE_DESC
= 'Youtube watch history, ":ythistory" for short (requires authentication)'
1584 _VALID_URL
= 'https?://www\.youtube\.com/feed/history|:ythistory'
1585 _FEED_NAME
= 'history'
1586 _PERSONAL_FEED
= True
1587 _PLAYLIST_TITLE
= 'Youtube Watch History'
1590 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
1591 IE_NAME
= 'youtube:favorites'
1592 IE_DESC
= 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
1593 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1594 _LOGIN_REQUIRED
= True
1596 def _real_extract(self
, url
):
1597 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1598 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, 'favourites playlist id')
1599 return self
.url_result(playlist_id
, 'YoutubePlaylist')
1602 class YoutubeSubscriptionsIE(YoutubePlaylistIE
):
1603 IE_NAME
= 'youtube:subscriptions'
1604 IE_DESC
= 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1605 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1608 def _real_extract(self
, url
):
1609 title
= 'Youtube Subscriptions'
1610 page
= self
._download
_webpage
('https://www.youtube.com/feed/subscriptions', title
)
1612 # The extraction process is the same as for playlists, but the regex
1613 # for the video ids doesn't contain an index
1615 more_widget_html
= content_html
= page
1617 for page_num
in itertools
.count(1):
1618 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
1619 new_ids
= orderedSet(matches
)
1622 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1626 more
= self
._download
_json
(
1627 'https://youtube.com/%s' % mobj
.group('more'), title
,
1628 'Downloading page #%s' % page_num
,
1629 transform_source
=uppercase_escape
)
1630 content_html
= more
['content_html']
1631 more_widget_html
= more
['load_more_widget_html']
1634 '_type': 'playlist',
1636 'entries': self
._ids
_to
_results
(ids
),
1640 class YoutubeTruncatedURLIE(InfoExtractor
):
1641 IE_NAME
= 'youtube:truncated_url'
1642 IE_DESC
= False # Do not list
1643 _VALID_URL
= r
'''(?x)
1644 (?:https?://)?[^/]+/watch\?(?:
1646 annotation_id=annotation_[^&]+
1648 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1652 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1653 'only_matching': True,
1655 'url': 'http://www.youtube.com/watch?',
1656 'only_matching': True,
1659 def _real_extract(self
, url
):
1660 raise ExtractorError(
1661 'Did you forget to quote the URL? Remember that & is a meta '
1662 'character in most shells, so you want to put the URL in quotes, '
1664 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1665 ' or simply youtube-dl BaW_jenozKc .',
1669 class YoutubeTruncatedIDIE(InfoExtractor
):
1670 IE_NAME
= 'youtube:truncated_id'
1671 IE_DESC
= False # Do not list
1672 _VALID_URL
= r
'https?://(?:www\.)youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
1675 'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
1676 'only_matching': True,
1679 def _real_extract(self
, url
):
1680 video_id
= self
._match
_id
(url
)
1681 raise ExtractorError(
1682 'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id
, url
),