3 from __future__
import unicode_literals
12 from .common
import InfoExtractor
, SearchInfoExtractor
13 from .subtitles
import SubtitlesInfoExtractor
14 from ..jsinterp
import JSInterpreter
15 from ..swfinterp
import SWFInterpreter
20 compat_urllib_request
,
26 get_element_by_attribute
,
36 class YoutubeBaseInfoExtractor(InfoExtractor
):
37 """Provide base functions for Youtube extractors"""
38 _LOGIN_URL
= 'https://accounts.google.com/ServiceLogin'
39 _TWOFACTOR_URL
= 'https://accounts.google.com/SecondFactor'
40 _LANG_URL
= r
'https://www.youtube.com/?hl=en&persist_hl=1&gl=US&persist_gl=1&opt_out_ackd=1'
41 _AGE_URL
= 'https://www.youtube.com/verify_age?next_url=/&gl=US&hl=en'
42 _NETRC_MACHINE
= 'youtube'
43 # If True it will raise an error if no login info is provided
44 _LOGIN_REQUIRED
= False
46 def _set_language(self
):
47 return bool(self
._download
_webpage
(
49 note
='Setting language', errnote
='unable to set language',
54 Attempt to log in to YouTube.
55 True is returned if successful or skipped.
56 False is returned if login failed.
58 If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
60 (username
, password
) = self
._get
_login
_info
()
61 # No authentication to be performed
63 if self
._LOGIN
_REQUIRED
:
64 raise ExtractorError('No login info available, needed for using %s.' % self
.IE_NAME
, expected
=True)
67 login_page
= self
._download
_webpage
(
68 self
._LOGIN
_URL
, None,
69 note
='Downloading login page',
70 errnote
='unable to fetch login page', fatal
=False)
71 if login_page
is False:
74 galx
= self
._search
_regex
(r
'(?s)<input.+?name="GALX".+?value="(.+?)"',
75 login_page
, 'Login GALX parameter')
79 'continue': 'https://www.youtube.com/signin?action_handle_signin=true&feature=sign_in_button&hl=en_US&nomobiletemp=1',
84 'PersistentCookie': 'yes',
86 'bgresponse': 'js_disabled',
87 'checkConnection': '',
88 'checkedDomains': 'youtube',
100 # Convert to UTF-8 *before* urlencode because Python 2.x's urlencode
102 login_form
= dict((k
.encode('utf-8'), v
.encode('utf-8')) for k
,v
in login_form_strs
.items())
103 login_data
= compat_urllib_parse
.urlencode(login_form
).encode('ascii')
105 req
= compat_urllib_request
.Request(self
._LOGIN
_URL
, login_data
)
106 login_results
= self
._download
_webpage
(
108 note
='Logging in', errnote
='unable to log in', fatal
=False)
109 if login_results
is False:
112 if re
.search(r
'id="errormsg_0_Passwd"', login_results
) is not None:
113 raise ExtractorError('Please use your account password and a two-factor code instead of an application-specific password.', expected
=True)
116 # TODO add SMS and phone call support - these require making a request and then prompting the user
118 if re
.search(r
'(?i)<form[^>]* id="gaia_secondfactorform"', login_results
) is not None:
119 tfa_code
= self
._get
_tfa
_info
()
122 self
._downloader
.report_warning('Two-factor authentication required. Provide it with --twofactor <code>')
123 self
._downloader
.report_warning('(Note that only TOTP (Google Authenticator App) codes work at this time.)')
126 # Unlike the first login form, secTok and timeStmp are both required for the TFA form
128 match
= re
.search(r
'id="secTok"\n\s+value=\'(.+)\'/>', login_results, re.M | re.U)
130 self._downloader.report_warning('Failed to get secTok
- did the page structure change?
')
131 secTok = match.group(1)
132 match = re.search(r'id="timeStmp"\n\s
+value
=\'(.+)\'/>', login_results, re.M | re.U)
134 self._downloader.report_warning('Failed to get timeStmp
- did the page structure change?
')
135 timeStmp = match.group(1)
138 'continue': 'https
://www
.youtube
.com
/signin?action_handle_signin
=true
&feature
=sign_in_button
&hl
=en_US
&nomobiletemp
=1',
140 'smsUserPin
': tfa_code,
141 'smsVerifyPin
': 'Verify
',
143 'PersistentCookie
': 'yes
',
144 'checkConnection
': '',
145 'checkedDomains
': 'youtube
',
148 'timeStmp
': timeStmp,
149 'service
': 'youtube
',
152 tfa_form = dict((k.encode('utf
-8'), v.encode('utf
-8')) for k,v in tfa_form_strs.items())
153 tfa_data = compat_urllib_parse.urlencode(tfa_form).encode('ascii
')
155 tfa_req = compat_urllib_request.Request(self._TWOFACTOR_URL, tfa_data)
156 tfa_results = self._download_webpage(
158 note='Submitting TFA code
', errnote='unable to submit tfa
', fatal=False)
160 if tfa_results is False:
163 if re.search(r'(?i
)<form
[^
>]* id="gaia_secondfactorform"', tfa_results) is not None:
164 self._downloader.report_warning('Two
-factor code expired
. Please
try again
, or use a one
-use backup code instead
.')
166 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', tfa_results) is not None:
167 self._downloader.report_warning('unable to log
in - did the page structure change?
')
169 if re.search(r'smsauth
-interstitial
-reviewsettings
', tfa_results) is not None:
170 self._downloader.report_warning('Your Google account has a security notice
. Please log
in on your web browser
, resolve the notice
, and try again
.')
173 if re.search(r'(?i
)<form
[^
>]* id="gaia_loginform"', login_results) is not None:
174 self._downloader.report_warning('unable to log
in: bad username
or password
')
178 def _confirm_age(self):
181 'action_confirm
': 'Confirm
',
183 req = compat_urllib_request.Request(self._AGE_URL,
184 compat_urllib_parse.urlencode(age_form).encode('ascii
'))
186 self._download_webpage(
188 note='Confirming age
', errnote='Unable to confirm age
',
191 def _real_initialize(self):
192 if self._downloader is None:
194 if self._get_login_info()[0] is not None:
195 if not self._set_language():
197 if not self._login():
202 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
203 IE_DESC = 'YouTube
.com
'
204 _VALID_URL = r"""(?x)^
206 (?:https?://|//) # http(s):// or protocol-independent URL
207 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
208 (?:www\.)?deturl\.com/www\.youtube\.com/|
209 (?:www\.)?pwnyoutube\.com/|
210 (?:www\.)?yourepeat\.com/|
211 tube\.majestyc\.net/|
212 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
213 (?:.*?\#/)? # handle anchor (#/) redirect urls
214 (?: # the various things that can precede the ID:
215 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
216 |(?: # or the v= param in all its forms
217 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
218 (?:\?|\#!?) # the params delimiter ? or # or #!
219 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
223 |youtu\.be/ # just youtu.be/xxxx
224 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
226 )? # all until now is optional -> you can pass the naked ID
227 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
228 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
229 (?(1).+)? # if we found the ID, everything can follow
231 _NEXT_URL_RE = r'[\?&]next_url
=([^
&]+)'
233 '5': {'ext': 'flv', 'width': 400, 'height': 240},
234 '6': {'ext': 'flv', 'width': 450, 'height': 270},
235 '13': {'ext': '3gp'},
236 '17': {'ext': '3gp', 'width': 176, 'height': 144},
237 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
238 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
239 '34': {'ext': 'flv', 'width': 640, 'height': 360},
240 '35': {'ext': 'flv', 'width': 854, 'height': 480},
241 '36': {'ext': '3gp', 'width': 320, 'height': 240},
242 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
243 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
244 '43': {'ext': 'webm', 'width': 640, 'height': 360},
245 '44': {'ext': 'webm', 'width': 854, 'height': 480},
246 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
247 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
251 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
252 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
253 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
254 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
255 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
256 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
257 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
259 # Apple HTTP Live Streaming
260 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
261 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
262 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
263 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
264 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
265 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
266 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
269 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
270 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
271 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
272 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
273 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
274 '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
275 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
276 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
277 '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
278 '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'h264'},
279 '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'vcodec': 'h264'},
282 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
283 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
284 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
287 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
288 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
289 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
290 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
291 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
292 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
293 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
294 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
298 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
299 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
300 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
301 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
302 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
303 '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
304 '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'fps': 60, 'vcodec': 'VP9'},
307 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
308 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
311 '_rtmp
': {'protocol': 'rtmp'},
317 'url
': 'http
://www
.youtube
.com
/watch?v
=BaW_jenozKc
',
321 'title
': 'youtube
-dl test video
"\'/\\ä↭𝕐',
322 'uploader': 'Philipp Hagemeister',
323 'uploader_id': 'phihag',
324 'upload_date': '20121002',
325 '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 .',
326 'categories
': ['Science
& Technology
'],
328 'dislike_count
': int,
332 'url
': 'http
://www
.youtube
.com
/watch?v
=UxxajLWwzqY
',
333 'note
': 'Test generic use_cipher_signature
video (#897)',
337 'upload_date': '20120506',
338 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
339 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
340 'uploader': 'Icona Pop',
341 'uploader_id': 'IconaPop',
345 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
346 'note': 'Test VEVO video with age protection (#956)',
350 'upload_date': '20130703',
351 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
352 'description': 'md5:64249768eec3bc4276236606ea996373',
353 'uploader': 'justintimberlakeVEVO',
354 'uploader_id': 'justintimberlakeVEVO',
358 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
359 'note': 'Embed-only video (#1746)',
363 'upload_date': '20120608',
364 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
365 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
366 'uploader': 'SET India',
367 'uploader_id': 'setindia'
371 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
372 'note': '256k DASH audio (format 141) via DASH manifest',
376 'upload_date': '20121002',
377 'uploader_id': '8KVIDEO',
379 'uploader': '8KVIDEO',
380 'title': 'UHDTV TEST 8K VIDEO.mp4'
383 'youtube_include_dash_manifest': True,
387 # DASH manifest with encrypted signature
389 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
393 'title': 'Afrojack - The Spark ft. Spree Wilson',
394 'description': 'md5:9717375db5a9a3992be4668bbf3bc0a8',
395 'uploader': 'AfrojackVEVO',
396 'uploader_id': 'AfrojackVEVO',
397 'upload_date': '20131011',
400 'youtube_include_dash_manifest': True,
406 def __init__(self
, *args
, **kwargs
):
407 super(YoutubeIE
, self
).__init
__(*args
, **kwargs
)
408 self
._player
_cache
= {}
410 def report_video_info_webpage_download(self
, video_id
):
411 """Report attempt to download video info webpage."""
412 self
.to_screen('%s: Downloading video info webpage' % video_id
)
414 def report_information_extraction(self
, video_id
):
415 """Report attempt to extract video information."""
416 self
.to_screen('%s: Extracting video information' % video_id
)
418 def report_unavailable_format(self
, video_id
, format
):
419 """Report extracted video URL."""
420 self
.to_screen('%s: Format %s not available' % (video_id
, format
))
422 def report_rtmp_download(self
):
423 """Indicate the download will use the RTMP protocol."""
424 self
.to_screen('RTMP download detected')
426 def _signature_cache_id(self
, example_sig
):
427 """ Return a string representation of a signature """
428 return '.'.join(compat_str(len(part
)) for part
in example_sig
.split('.'))
430 def _extract_signature_function(self
, video_id
, player_url
, example_sig
):
432 r
'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
435 raise ExtractorError('Cannot identify player %r' % player_url
)
436 player_type
= id_m
.group('ext')
437 player_id
= id_m
.group('id')
439 # Read from filesystem cache
440 func_id
= '%s_%s_%s' % (
441 player_type
, player_id
, self
._signature
_cache
_id
(example_sig
))
442 assert os
.path
.basename(func_id
) == func_id
444 cache_spec
= self
._downloader
.cache
.load('youtube-sigfuncs', func_id
)
445 if cache_spec
is not None:
446 return lambda s
: ''.join(s
[i
] for i
in cache_spec
)
448 if player_type
== 'js':
449 code
= self
._download
_webpage
(
450 player_url
, video_id
,
451 note
='Downloading %s player %s' % (player_type
, player_id
),
452 errnote
='Download of %s failed' % player_url
)
453 res
= self
._parse
_sig
_js
(code
)
454 elif player_type
== 'swf':
455 urlh
= self
._request
_webpage
(
456 player_url
, video_id
,
457 note
='Downloading %s player %s' % (player_type
, player_id
),
458 errnote
='Download of %s failed' % player_url
)
460 res
= self
._parse
_sig
_swf
(code
)
462 assert False, 'Invalid player type %r' % player_type
464 if cache_spec
is None:
465 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
466 cache_res
= res(test_string
)
467 cache_spec
= [ord(c
) for c
in cache_res
]
469 self
._downloader
.cache
.store('youtube-sigfuncs', func_id
, cache_spec
)
472 def _print_sig_code(self
, func
, example_sig
):
473 def gen_sig_code(idxs
):
474 def _genslice(start
, end
, step
):
475 starts
= '' if start
== 0 else str(start
)
476 ends
= (':%d' % (end
+step
)) if end
+ step
>= 0 else ':'
477 steps
= '' if step
== 1 else (':%d' % step
)
478 return 's[%s%s%s]' % (starts
, ends
, steps
)
481 start
= '(Never used)' # Quelch pyflakes warnings - start will be
482 # set as soon as step is set
483 for i
, prev
in zip(idxs
[1:], idxs
[:-1]):
487 yield _genslice(start
, prev
, step
)
490 if i
- prev
in [-1, 1]:
499 yield _genslice(start
, i
, step
)
501 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
502 cache_res
= func(test_string
)
503 cache_spec
= [ord(c
) for c
in cache_res
]
504 expr_code
= ' + '.join(gen_sig_code(cache_spec
))
505 signature_id_tuple
= '(%s)' % (
506 ', '.join(compat_str(len(p
)) for p
in example_sig
.split('.')))
507 code
= ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
508 ' return %s\n') % (signature_id_tuple
, expr_code
)
509 self
.to_screen('Extracted signature function:\n' + code
)
511 def _parse_sig_js(self
, jscode
):
512 funcname
= self
._search
_regex
(
513 r
'signature=([$a-zA-Z]+)', jscode
,
514 'Initial JS player signature function name')
516 jsi
= JSInterpreter(jscode
)
517 initial_function
= jsi
.extract_function(funcname
)
518 return lambda s
: initial_function([s
])
520 def _parse_sig_swf(self
, file_contents
):
521 swfi
= SWFInterpreter(file_contents
)
522 TARGET_CLASSNAME
= 'SignatureDecipher'
523 searched_class
= swfi
.extract_class(TARGET_CLASSNAME
)
524 initial_function
= swfi
.extract_function(searched_class
, 'decipher')
525 return lambda s
: initial_function([s
])
527 def _decrypt_signature(self
, s
, video_id
, player_url
, age_gate
=False):
528 """Turn the encrypted s field into a working signature"""
530 if player_url
is None:
531 raise ExtractorError('Cannot decrypt signature without player_url')
533 if player_url
.startswith('//'):
534 player_url
= 'https:' + player_url
536 player_id
= (player_url
, self
._signature
_cache
_id
(s
))
537 if player_id
not in self
._player
_cache
:
538 func
= self
._extract
_signature
_function
(
539 video_id
, player_url
, s
541 self
._player
_cache
[player_id
] = func
542 func
= self
._player
_cache
[player_id
]
543 if self
._downloader
.params
.get('youtube_print_sig_code'):
544 self
._print
_sig
_code
(func
, s
)
546 except Exception as e
:
547 tb
= traceback
.format_exc()
548 raise ExtractorError(
549 'Signature extraction failed: ' + tb
, cause
=e
)
551 def _get_available_subtitles(self
, video_id
, webpage
):
553 sub_list
= self
._download
_webpage
(
554 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
,
555 video_id
, note
=False)
556 except ExtractorError
as err
:
557 self
._downloader
.report_warning('unable to download video subtitles: %s' % compat_str(err
))
559 lang_list
= re
.findall(r
'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list
)
564 if lang
in sub_lang_list
:
566 params
= compat_urllib_parse
.urlencode({
569 'fmt': self
._downloader
.params
.get('subtitlesformat', 'srt'),
570 'name': unescapeHTML(l
[0]).encode('utf-8'),
572 url
= 'https://www.youtube.com/api/timedtext?' + params
573 sub_lang_list
[lang
] = url
574 if not sub_lang_list
:
575 self
._downloader
.report_warning('video doesn\'t have subtitles')
579 def _get_available_automatic_caption(self
, video_id
, webpage
):
580 """We need the webpage for getting the captions url, pass it as an
581 argument to speed up the process."""
582 sub_format
= self
._downloader
.params
.get('subtitlesformat', 'srt')
583 self
.to_screen('%s: Looking for automatic captions' % video_id
)
584 mobj
= re
.search(r
';ytplayer.config = ({.*?});', webpage
)
585 err_msg
= 'Couldn\'t find automatic captions for %s' % video_id
587 self
._downloader
.report_warning(err_msg
)
589 player_config
= json
.loads(mobj
.group(1))
591 args
= player_config
[u
'args']
592 caption_url
= args
[u
'ttsurl']
593 timestamp
= args
[u
'timestamp']
594 # We get the available subtitles
595 list_params
= compat_urllib_parse
.urlencode({
600 list_url
= caption_url
+ '&' + list_params
601 caption_list
= self
._download
_xml
(list_url
, video_id
)
602 original_lang_node
= caption_list
.find('track')
603 if original_lang_node
is None or original_lang_node
.attrib
.get('kind') != 'asr' :
604 self
._downloader
.report_warning('Video doesn\'t have automatic captions')
606 original_lang
= original_lang_node
.attrib
['lang_code']
609 for lang_node
in caption_list
.findall('target'):
610 sub_lang
= lang_node
.attrib
['lang_code']
611 params
= compat_urllib_parse
.urlencode({
612 'lang': original_lang
,
618 sub_lang_list
[sub_lang
] = caption_url
+ '&' + params
620 # An extractor error can be raise by the download process if there are
621 # no automatic captions but there are subtitles
622 except (KeyError, ExtractorError
):
623 self
._downloader
.report_warning(err_msg
)
627 def extract_id(cls
, url
):
628 mobj
= re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
)
630 raise ExtractorError('Invalid URL: %s' % url
)
631 video_id
= mobj
.group(2)
634 def _extract_from_m3u8(self
, manifest_url
, video_id
):
636 def _get_urls(_manifest
):
637 lines
= _manifest
.split('\n')
638 urls
= filter(lambda l
: l
and not l
.startswith('#'),
641 manifest
= self
._download
_webpage
(manifest_url
, video_id
, 'Downloading formats manifest')
642 formats_urls
= _get_urls(manifest
)
643 for format_url
in formats_urls
:
644 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
645 url_map
[itag
] = format_url
648 def _extract_annotations(self
, video_id
):
649 url
= 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
650 return self
._download
_webpage
(url
, video_id
, note
='Searching for annotations.', errnote
='Unable to download video annotations.')
652 def _real_extract(self
, url
):
654 'http' if self
._downloader
.params
.get('prefer_insecure', False)
657 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
658 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
660 url
= proto
+ '://www.youtube.com/' + compat_urllib_parse
.unquote(mobj
.group(1)).lstrip('/')
661 video_id
= self
.extract_id(url
)
664 url
= proto
+ '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
666 c
for c
in self
._downloader
.cookiejar
667 if c
.domain
== '.youtube.com' and c
.name
== 'PREF']
668 for pc
in pref_cookies
:
669 if 'hl=' in pc
.value
:
670 pc
.value
= re
.sub(r
'hl=[^&]+', 'hl=en', pc
.value
)
675 video_webpage
= self
._download
_webpage
(url
, video_id
)
677 # Attempt to extract SWF player URL
678 mobj
= re
.search(r
'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
680 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
685 self
.report_video_info_webpage_download(video_id
)
686 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
688 # We simulate the access to the video from www.youtube.com/v/{video_id}
689 # this can be viewed without login into Youtube
690 data
= compat_urllib_parse
.urlencode({
691 'video_id': video_id
,
692 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
693 'sts': self
._search
_regex
(
694 r
'"sts"\s*:\s*(\d+)', video_webpage
, 'sts', default
=''),
696 video_info_url
= proto
+ '://www.youtube.com/get_video_info?' + data
697 video_info_webpage
= self
._download
_webpage
(
698 video_info_url
, video_id
,
699 note
='Refetching age-gated info webpage',
700 errnote
='unable to download video info webpage')
701 video_info
= compat_parse_qs(video_info_webpage
)
704 for el_type
in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
705 video_info_url
= (proto
+ '://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
706 % (video_id
, el_type
))
707 video_info_webpage
= self
._download
_webpage
(video_info_url
, video_id
,
709 errnote
='unable to download video info webpage')
710 video_info
= compat_parse_qs(video_info_webpage
)
711 if 'token' in video_info
:
713 if 'token' not in video_info
:
714 if 'reason' in video_info
:
715 raise ExtractorError(
716 'YouTube said: %s' % video_info
['reason'][0],
717 expected
=True, video_id
=video_id
)
719 raise ExtractorError(
720 '"token" parameter not in video info for unknown reason',
723 if 'view_count' in video_info
:
724 view_count
= int(video_info
['view_count'][0])
728 # Check for "rental" videos
729 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
730 raise ExtractorError('"rental" videos not supported')
732 # Start extracting information
733 self
.report_information_extraction(video_id
)
736 if 'author' not in video_info
:
737 raise ExtractorError('Unable to extract uploader name')
738 video_uploader
= compat_urllib_parse
.unquote_plus(video_info
['author'][0])
741 video_uploader_id
= None
742 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
744 video_uploader_id
= mobj
.group(1)
746 self
._downloader
.report_warning('unable to extract uploader nickname')
749 if 'title' in video_info
:
750 video_title
= video_info
['title'][0]
752 self
._downloader
.report_warning('Unable to extract video title')
756 # We try first to get a high quality image:
757 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
758 video_webpage
, re
.DOTALL
)
759 if m_thumb
is not None:
760 video_thumbnail
= m_thumb
.group(1)
761 elif 'thumbnail_url' not in video_info
:
762 self
._downloader
.report_warning('unable to extract video thumbnail')
763 video_thumbnail
= None
764 else: # don't panic if we can't find it
765 video_thumbnail
= compat_urllib_parse
.unquote_plus(video_info
['thumbnail_url'][0])
769 mobj
= re
.search(r
'(?s)id="eow-date.*?>(.*?)</span>', video_webpage
)
772 r
'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
775 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
776 upload_date
= unified_strdate(upload_date
)
778 m_cat_container
= self
._search
_regex
(
779 r
'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
780 video_webpage
, 'categories', fatal
=False)
782 category
= self
._html
_search
_regex
(
783 r
'(?s)<a[^<]+>(.*?)</a>', m_cat_container
, 'category',
785 video_categories
= None if category
is None else [category
]
787 video_categories
= None
790 video_description
= get_element_by_id("eow-description", video_webpage
)
791 if video_description
:
792 video_description
= re
.sub(r
'''(?x)
794 (?:[a-zA-Z-]+="[^"]+"\s+)*?
796 (?:[a-zA-Z-]+="[^"]+"\s+)*?
797 class="yt-uix-redirect-link"\s*>
800 ''', r
'\1', video_description
)
801 video_description
= clean_html(video_description
)
803 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
805 video_description
= unescapeHTML(fd_mobj
.group(1))
807 video_description
= ''
809 def _extract_count(count_name
):
810 count
= self
._search
_regex
(
811 r
'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re
.escape(count_name
),
812 video_webpage
, count_name
, default
=None)
813 if count
is not None:
814 return int(count
.replace(',', ''))
816 like_count
= _extract_count('like')
817 dislike_count
= _extract_count('dislike')
820 video_subtitles
= self
.extract_subtitles(video_id
, video_webpage
)
822 if self
._downloader
.params
.get('listsubtitles', False):
823 self
._list
_available
_subtitles
(video_id
, video_webpage
)
826 if 'length_seconds' not in video_info
:
827 self
._downloader
.report_warning('unable to extract video duration')
828 video_duration
= None
830 video_duration
= int(compat_urllib_parse
.unquote_plus(video_info
['length_seconds'][0]))
833 video_annotations
= None
834 if self
._downloader
.params
.get('writeannotations', False):
835 video_annotations
= self
._extract
_annotations
(video_id
)
837 # Decide which formats to download
839 mobj
= re
.search(r
';ytplayer\.config\s*=\s*({.*?});', video_webpage
)
841 raise ValueError('Could not find vevo ID')
842 json_code
= uppercase_escape(mobj
.group(1))
843 ytplayer_config
= json
.loads(json_code
)
844 args
= ytplayer_config
['args']
845 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
846 # this signatures are encrypted
847 if 'url_encoded_fmt_stream_map' not in args
:
848 raise ValueError('No stream_map present') # caught below
849 re_signature
= re
.compile(r
'[&,]s=')
850 m_s
= re_signature
.search(args
['url_encoded_fmt_stream_map'])
852 self
.to_screen('%s: Encrypted signatures detected.' % video_id
)
853 video_info
['url_encoded_fmt_stream_map'] = [args
['url_encoded_fmt_stream_map']]
854 m_s
= re_signature
.search(args
.get('adaptive_fmts', ''))
856 if 'adaptive_fmts' in video_info
:
857 video_info
['adaptive_fmts'][0] += ',' + args
['adaptive_fmts']
859 video_info
['adaptive_fmts'] = [args
['adaptive_fmts']]
863 def _map_to_format_list(urlmap
):
865 for itag
, video_real_url
in urlmap
.items():
868 'url': video_real_url
,
869 'player_url': player_url
,
871 if itag
in self
._formats
:
872 dct
.update(self
._formats
[itag
])
876 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
877 self
.report_rtmp_download()
879 'format_id': '_rtmp',
881 'url': video_info
['conn'][0],
882 'player_url': player_url
,
884 elif len(video_info
.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info
.get('adaptive_fmts', [])) >= 1:
885 encoded_url_map
= video_info
.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info
.get('adaptive_fmts',[''])[0]
886 if 'rtmpe%3Dyes' in encoded_url_map
:
887 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
889 for url_data_str
in encoded_url_map
.split(','):
890 url_data
= compat_parse_qs(url_data_str
)
891 if 'itag' not in url_data
or 'url' not in url_data
:
893 format_id
= url_data
['itag'][0]
894 url
= url_data
['url'][0]
896 if 'sig' in url_data
:
897 url
+= '&signature=' + url_data
['sig'][0]
898 elif 's' in url_data
:
899 encrypted_sig
= url_data
['s'][0]
902 jsplayer_url_json
= self
._search
_regex
(
903 r
'"assets":.+?"js":\s*("[^"]+")',
904 video_webpage
, 'JS player URL')
905 player_url
= json
.loads(jsplayer_url_json
)
906 if player_url
is None:
907 player_url_json
= self
._search
_regex
(
908 r
'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
909 video_webpage
, 'age gate player URL')
910 player_url
= json
.loads(player_url_json
)
912 if self
._downloader
.params
.get('verbose'):
913 if player_url
is None:
914 player_version
= 'unknown'
915 player_desc
= 'unknown'
917 if player_url
.endswith('swf'):
918 player_version
= self
._search
_regex
(
919 r
'-(.+?)(?:/watch_as3)?\.swf$', player_url
,
920 'flash player', fatal
=False)
921 player_desc
= 'flash player %s' % player_version
923 player_version
= self
._search
_regex
(
924 r
'html5player-([^/]+?)(?:/html5player)?\.js',
926 'html5 player', fatal
=False)
927 player_desc
= 'html5 player %s' % player_version
929 parts_sizes
= self
._signature
_cache
_id
(encrypted_sig
)
930 self
.to_screen('{%s} signature length %s, %s' %
931 (format_id
, parts_sizes
, player_desc
))
933 signature
= self
._decrypt
_signature
(
934 encrypted_sig
, video_id
, player_url
, age_gate
)
935 url
+= '&signature=' + signature
936 if 'ratebypass' not in url
:
937 url
+= '&ratebypass=yes'
938 url_map
[format_id
] = url
939 formats
= _map_to_format_list(url_map
)
940 elif video_info
.get('hlsvp'):
941 manifest_url
= video_info
['hlsvp'][0]
942 url_map
= self
._extract
_from
_m
3u8(manifest_url
, video_id
)
943 formats
= _map_to_format_list(url_map
)
945 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
947 # Look for the DASH manifest
948 if self
._downloader
.params
.get('youtube_include_dash_manifest', True):
950 # The DASH manifest used needs to be the one from the original video_webpage.
951 # The one found in get_video_info seems to be using different signatures.
952 # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
953 # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
954 # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
956 dash_manifest_url
= video_info
.get('dashmpd')[0]
958 dash_manifest_url
= ytplayer_config
['args']['dashmpd']
959 def decrypt_sig(mobj
):
961 dec_s
= self
._decrypt
_signature
(s
, video_id
, player_url
, age_gate
)
962 return '/signature/%s' % dec_s
963 dash_manifest_url
= re
.sub(r
'/s/([\w\.]+)', decrypt_sig
, dash_manifest_url
)
964 dash_doc
= self
._download
_xml
(
965 dash_manifest_url
, video_id
,
966 note
='Downloading DASH manifest',
967 errnote
='Could not download DASH manifest')
968 for r
in dash_doc
.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
969 url_el
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
972 format_id
= r
.attrib
['id']
973 video_url
= url_el
.text
974 filesize
= int_or_none(url_el
.attrib
.get('{http://youtube.com/yt/2012/10/10}contentLength'))
976 'format_id': format_id
,
978 'width': int_or_none(r
.attrib
.get('width')),
979 'tbr': int_or_none(r
.attrib
.get('bandwidth'), 1000),
980 'asr': int_or_none(r
.attrib
.get('audioSamplingRate')),
981 'filesize': filesize
,
984 existing_format
= next(
986 if fo
['format_id'] == format_id
)
987 except StopIteration:
988 f
.update(self
._formats
.get(format_id
, {}))
991 existing_format
.update(f
)
993 except (ExtractorError
, KeyError) as e
:
994 self
.report_warning('Skipping DASH manifest: %s' % e
, video_id
)
996 self
._sort
_formats
(formats
)
1000 'uploader': video_uploader
,
1001 'uploader_id': video_uploader_id
,
1002 'upload_date': upload_date
,
1003 'title': video_title
,
1004 'thumbnail': video_thumbnail
,
1005 'description': video_description
,
1006 'categories': video_categories
,
1007 'subtitles': video_subtitles
,
1008 'duration': video_duration
,
1009 'age_limit': 18 if age_gate
else 0,
1010 'annotations': video_annotations
,
1011 'webpage_url': proto
+ '://www.youtube.com/watch?v=%s' % video_id
,
1012 'view_count': view_count
,
1013 'like_count': like_count
,
1014 'dislike_count': dislike_count
,
1018 class YoutubePlaylistIE(YoutubeBaseInfoExtractor
):
1019 IE_DESC
= 'YouTube.com playlists'
1020 _VALID_URL
= r
"""(?x)(?:
1025 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1026 \? (?:.*?&)*? (?:p|a|list)=
1030 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1031 # Top tracks, they can also include dots
1036 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1038 _TEMPLATE_URL
= 'https://www.youtube.com/playlist?list=%s'
1039 _MORE_PAGES_INDICATOR
= r
'data-link-type="next"'
1040 _VIDEO_RE
= r
'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1041 IE_NAME
= 'youtube:playlist'
1043 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1045 'title': 'ytdl test PL',
1047 'playlist_count': 3,
1049 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1051 'title': 'YDL_Empty_List',
1053 'playlist_count': 0,
1055 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1056 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1058 'title': '29C3: Not my department',
1060 'playlist_count': 95,
1062 'note': 'issue #673',
1063 'url': 'PLBB231211A4F62143',
1065 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1067 'playlist_mincount': 26,
1069 'note': 'Large playlist',
1070 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1072 'title': 'Uploads from Cauchemar',
1074 'playlist_mincount': 799,
1076 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1078 'title': 'YDL_safe_search',
1080 'playlist_count': 2,
1083 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1084 'playlist_count': 4,
1089 'note': 'Embedded SWF player',
1090 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1091 'playlist_count': 4,
1097 def _real_initialize(self
):
1100 def _ids_to_results(self
, ids
):
1102 self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
1105 def _extract_mix(self
, playlist_id
):
1106 # The mixes are generated from a a single video
1107 # the id of the playlist is just 'RD' + video_id
1108 url
= 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id
[-11:], playlist_id
)
1109 webpage
= self
._download
_webpage
(
1110 url
, playlist_id
, 'Downloading Youtube mix')
1111 search_title
= lambda class_name
: get_element_by_attribute('class', class_name
, webpage
)
1113 search_title('playlist-title') or
1114 search_title('title long-title') or
1115 search_title('title'))
1116 title
= clean_html(title_span
)
1117 ids
= orderedSet(re
.findall(
1118 r
'''(?xs)data-video-username=".*?".*?
1119 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re
.escape(playlist_id
),
1121 url_results
= self
._ids
_to
_results
(ids
)
1123 return self
.playlist_result(url_results
, playlist_id
, title
)
1125 def _real_extract(self
, url
):
1126 # Extract playlist id
1127 mobj
= re
.match(self
._VALID
_URL
, url
)
1129 raise ExtractorError('Invalid URL: %s' % url
)
1130 playlist_id
= mobj
.group(1) or mobj
.group(2)
1132 # Check if it's a video-specific URL
1133 query_dict
= compat_urlparse
.parse_qs(compat_urlparse
.urlparse(url
).query
)
1134 if 'v' in query_dict
:
1135 video_id
= query_dict
['v'][0]
1136 if self
._downloader
.params
.get('noplaylist'):
1137 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1138 return self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1140 self
.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id
, video_id
))
1142 if playlist_id
.startswith('RD'):
1143 # Mixes require a custom extraction process
1144 return self
._extract
_mix
(playlist_id
)
1145 if playlist_id
.startswith('TL'):
1146 raise ExtractorError('For downloading YouTube.com top lists, use '
1147 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected
=True)
1149 url
= self
._TEMPLATE
_URL
% playlist_id
1150 page
= self
._download
_webpage
(url
, playlist_id
)
1151 more_widget_html
= content_html
= page
1153 # Check if the playlist exists or is private
1154 if re
.search(r
'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page
) is not None:
1155 raise ExtractorError(
1156 'The playlist doesn\'t exist or is private, use --username or '
1157 '--netrc to access it.',
1160 # Extract the video ids from the playlist pages
1163 for page_num
in itertools
.count(1):
1164 matches
= re
.finditer(self
._VIDEO
_RE
, content_html
)
1165 # We remove the duplicates and the link with index 0
1166 # (it's not the first video of the playlist)
1167 new_ids
= orderedSet(m
.group('id') for m
in matches
if m
.group('index') != '0')
1170 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1174 more
= self
._download
_json
(
1175 'https://youtube.com/%s' % mobj
.group('more'), playlist_id
,
1176 'Downloading page #%s' % page_num
,
1177 transform_source
=uppercase_escape
)
1178 content_html
= more
['content_html']
1179 more_widget_html
= more
['load_more_widget_html']
1181 playlist_title
= self
._html
_search
_regex
(
1182 r
'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1185 url_results
= self
._ids
_to
_results
(ids
)
1186 return self
.playlist_result(url_results
, playlist_id
, playlist_title
)
1189 class YoutubeTopListIE(YoutubePlaylistIE
):
1190 IE_NAME
= 'youtube:toplist'
1191 IE_DESC
= ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1192 ' (Example: "yttoplist:music:Top Tracks")')
1193 _VALID_URL
= r
'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1195 'url': 'yttoplist:music:Trending',
1196 'playlist_mincount': 5,
1197 'skip': 'Only works for logged-in users',
1200 def _real_extract(self
, url
):
1201 mobj
= re
.match(self
._VALID
_URL
, url
)
1202 channel
= mobj
.group('chann')
1203 title
= mobj
.group('title')
1204 query
= compat_urllib_parse
.urlencode({'title': title}
)
1205 channel_page
= self
._download
_webpage
(
1206 'https://www.youtube.com/%s' % channel
, title
)
1207 link
= self
._html
_search
_regex
(
1209 <a\s+href="([^"]+)".*?>\s*
1210 <span\s+class="branded-page-module-title-text">\s*
1211 <span[^>]*>.*?%s.*?</span>''' % re
.escape(query
),
1212 channel_page
, 'list')
1213 url
= compat_urlparse
.urljoin('https://www.youtube.com/', link
)
1215 video_re
= r
'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1217 # sometimes the webpage doesn't contain the videos
1218 # retry until we get them
1219 for i
in itertools
.count(0):
1220 msg
= 'Downloading Youtube mix'
1222 msg
+= ', retry #%d' % i
1224 webpage
= self
._download
_webpage
(url
, title
, msg
)
1225 ids
= orderedSet(re
.findall(video_re
, webpage
))
1228 url_results
= self
._ids
_to
_results
(ids
)
1229 return self
.playlist_result(url_results
, playlist_title
=title
)
1232 class YoutubeChannelIE(InfoExtractor
):
1233 IE_DESC
= 'YouTube.com channels'
1234 _VALID_URL
= r
"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
1235 _MORE_PAGES_INDICATOR
= 'yt-uix-load-more'
1236 _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'
1237 IE_NAME
= 'youtube:channel'
1239 'note': 'paginated channel',
1240 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1241 'playlist_mincount': 91,
1244 def extract_videos_from_page(self
, page
):
1246 for mobj
in re
.finditer(r
'href="/watch\?v=([0-9A-Za-z_-]+)&?', page
):
1247 if mobj
.group(1) not in ids_in_page
:
1248 ids_in_page
.append(mobj
.group(1))
1251 def _real_extract(self
, url
):
1252 # Extract channel id
1253 mobj
= re
.match(self
._VALID
_URL
, url
)
1255 raise ExtractorError('Invalid URL: %s' % url
)
1257 # Download channel page
1258 channel_id
= mobj
.group(1)
1260 url
= 'https://www.youtube.com/channel/%s/videos' % channel_id
1261 channel_page
= self
._download
_webpage
(url
, channel_id
)
1262 autogenerated
= re
.search(r
'''(?x)
1264 channel-header-autogenerated-label|
1265 yt-channel-title-autogenerated
1266 )[^"]*"''', channel_page
) is not None
1269 # The videos are contained in a single page
1270 # the ajax pages can't be used, they are empty
1271 video_ids
= self
.extract_videos_from_page(channel_page
)
1273 # Download all channel pages using the json-based channel_ajax query
1274 for pagenum
in itertools
.count(1):
1275 url
= self
._MORE
_PAGES
_URL
% (pagenum
, channel_id
)
1276 page
= self
._download
_json
(
1277 url
, channel_id
, note
='Downloading page #%s' % pagenum
,
1278 transform_source
=uppercase_escape
)
1280 ids_in_page
= self
.extract_videos_from_page(page
['content_html'])
1281 video_ids
.extend(ids_in_page
)
1283 if self
._MORE
_PAGES
_INDICATOR
not in page
['load_more_widget_html']:
1286 self
._downloader
.to_screen('[youtube] Channel %s: Found %i videos' % (channel_id
, len(video_ids
)))
1288 url_entries
= [self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1289 for video_id
in video_ids
]
1290 return self
.playlist_result(url_entries
, channel_id
)
1293 class YoutubeUserIE(InfoExtractor
):
1294 IE_DESC
= 'YouTube.com user videos (URL or "ytuser" keyword)'
1295 _VALID_URL
= r
'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
1296 _TEMPLATE_URL
= 'https://gdata.youtube.com/feeds/api/users/%s'
1297 _GDATA_PAGE_SIZE
= 50
1298 _GDATA_URL
= 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1299 IE_NAME
= 'youtube:user'
1302 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1303 'playlist_mincount': 320,
1305 'title': 'TheLinuxFoundation',
1308 'url': 'ytuser:phihag',
1309 'only_matching': True,
1313 def suitable(cls
, url
):
1314 # Don't return True if the url can be extracted with other youtube
1315 # extractor, the regex would is too permissive and it would match.
1316 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1317 if any(ie
.suitable(url
) for ie
in other_ies
): return False
1318 else: return super(YoutubeUserIE
, cls
).suitable(url
)
1320 def _real_extract(self
, url
):
1322 mobj
= re
.match(self
._VALID
_URL
, url
)
1324 raise ExtractorError('Invalid URL: %s' % url
)
1326 username
= mobj
.group(1)
1328 # Download video ids using YouTube Data API. Result size per
1329 # query is limited (currently to 50 videos) so we need to query
1330 # page by page until there are no video ids - it means we got
1333 def download_page(pagenum
):
1334 start_index
= pagenum
* self
._GDATA
_PAGE
_SIZE
+ 1
1336 gdata_url
= self
._GDATA
_URL
% (username
, self
._GDATA
_PAGE
_SIZE
, start_index
)
1337 page
= self
._download
_webpage
(
1338 gdata_url
, username
,
1339 'Downloading video ids from %d to %d' % (
1340 start_index
, start_index
+ self
._GDATA
_PAGE
_SIZE
))
1343 response
= json
.loads(page
)
1344 except ValueError as err
:
1345 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err
))
1346 if 'entry' not in response
['feed']:
1349 # Extract video identifiers
1350 entries
= response
['feed']['entry']
1351 for entry
in entries
:
1352 title
= entry
['title']['$t']
1353 video_id
= entry
['id']['$t'].split('/')[-1]
1357 'ie_key': 'Youtube',
1361 url_results
= OnDemandPagedList(download_page
, self
._GDATA
_PAGE
_SIZE
)
1363 return self
.playlist_result(url_results
, playlist_title
=username
)
1366 class YoutubeSearchIE(SearchInfoExtractor
):
1367 IE_DESC
= 'YouTube.com searches'
1368 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1370 IE_NAME
= 'youtube:search'
1371 _SEARCH_KEY
= 'ytsearch'
1373 def _get_n_results(self
, query
, n
):
1374 """Get a specified number of results for a query"""
1381 while (PAGE_SIZE
* pagenum
) < limit
:
1382 result_url
= self
._API
_URL
% (
1383 compat_urllib_parse
.quote_plus(query
.encode('utf-8')),
1384 (PAGE_SIZE
* pagenum
) + 1)
1385 data_json
= self
._download
_webpage
(
1386 result_url
, video_id
='query "%s"' % query
,
1387 note
='Downloading page %s' % (pagenum
+ 1),
1388 errnote
='Unable to download API page')
1389 data
= json
.loads(data_json
)
1390 api_response
= data
['data']
1392 if 'items' not in api_response
:
1393 raise ExtractorError(
1394 '[youtube] No video results', expected
=True)
1396 new_ids
= list(video
['id'] for video
in api_response
['items'])
1397 video_ids
+= new_ids
1399 limit
= min(n
, api_response
['totalItems'])
1402 if len(video_ids
) > n
:
1403 video_ids
= video_ids
[:n
]
1404 videos
= [self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1405 for video_id
in video_ids
]
1406 return self
.playlist_result(videos
, query
)
1409 class YoutubeSearchDateIE(YoutubeSearchIE
):
1410 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1411 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1412 _SEARCH_KEY
= 'ytsearchdate'
1413 IE_DESC
= 'YouTube.com searches, newest videos first'
1416 class YoutubeSearchURLIE(InfoExtractor
):
1417 IE_DESC
= 'YouTube.com search URLs'
1418 IE_NAME
= 'youtube:search_url'
1419 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1421 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1422 'playlist_mincount': 5,
1424 'title': 'youtube-dl test video',
1428 def _real_extract(self
, url
):
1429 mobj
= re
.match(self
._VALID
_URL
, url
)
1430 query
= compat_urllib_parse
.unquote_plus(mobj
.group('query'))
1432 webpage
= self
._download
_webpage
(url
, query
)
1433 result_code
= self
._search
_regex
(
1434 r
'(?s)<ol class="item-section"(.*?)</ol>', webpage
, 'result HTML')
1436 part_codes
= re
.findall(
1437 r
'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code
)
1439 for part_code
in part_codes
:
1440 part_title
= self
._html
_search
_regex
(
1441 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1442 part_url_snippet
= self
._html
_search
_regex
(
1443 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1444 part_url
= compat_urlparse
.urljoin(
1445 'https://www.youtube.com/', part_url_snippet
)
1449 'title': part_title
,
1453 '_type': 'playlist',
1459 class YoutubeShowIE(InfoExtractor
):
1460 IE_DESC
= 'YouTube.com (multi-season) shows'
1461 _VALID_URL
= r
'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1462 IE_NAME
= 'youtube:show'
1464 'url': 'http://www.youtube.com/show/airdisasters',
1465 'playlist_mincount': 3,
1467 'id': 'airdisasters',
1468 'title': 'Air Disasters',
1472 def _real_extract(self
, url
):
1473 mobj
= re
.match(self
._VALID
_URL
, url
)
1474 playlist_id
= mobj
.group('id')
1475 webpage
= self
._download
_webpage
(
1476 url
, playlist_id
, 'Downloading show webpage')
1477 # There's one playlist for each season of the show
1478 m_seasons
= list(re
.finditer(r
'href="(/playlist\?list=.*?)"', webpage
))
1479 self
.to_screen('%s: Found %s seasons' % (playlist_id
, len(m_seasons
)))
1482 'https://www.youtube.com' + season
.group(1), 'YoutubePlaylist')
1483 for season
in m_seasons
1485 title
= self
._og
_search
_title
(webpage
, fatal
=False)
1488 '_type': 'playlist',
1495 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1497 Base class for extractors that fetch info from
1498 http://www.youtube.com/feed_ajax
1499 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1501 _LOGIN_REQUIRED
= True
1502 # use action_load_personal_feed instead of action_load_system_feed
1503 _PERSONAL_FEED
= False
1506 def _FEED_TEMPLATE(self
):
1507 action
= 'action_load_system_feed'
1508 if self
._PERSONAL
_FEED
:
1509 action
= 'action_load_personal_feed'
1510 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action
, self
._FEED
_NAME
)
1514 return 'youtube:%s' % self
._FEED
_NAME
1516 def _real_initialize(self
):
1519 def _real_extract(self
, url
):
1522 for i
in itertools
.count(1):
1523 info
= self
._download
_json
(self
._FEED
_TEMPLATE
% paging
,
1524 '%s feed' % self
._FEED
_NAME
,
1525 'Downloading page %s' % i
)
1526 feed_html
= info
.get('feed_html') or info
.get('content_html')
1527 load_more_widget_html
= info
.get('load_more_widget_html') or feed_html
1528 m_ids
= re
.finditer(r
'"/watch\?v=(.*?)["&]', feed_html
)
1529 ids
= orderedSet(m
.group(1) for m
in m_ids
)
1530 feed_entries
.extend(
1531 self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1532 for video_id
in ids
)
1534 r
'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1535 load_more_widget_html
)
1538 paging
= mobj
.group('paging')
1539 return self
.playlist_result(feed_entries
, playlist_title
=self
._PLAYLIST
_TITLE
)
1541 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
1542 IE_DESC
= 'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1543 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1544 _FEED_NAME
= 'recommended'
1545 _PLAYLIST_TITLE
= 'Youtube Recommended videos'
1547 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor
):
1548 IE_DESC
= 'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1549 _VALID_URL
= r
'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1550 _FEED_NAME
= 'watch_later'
1551 _PLAYLIST_TITLE
= 'Youtube Watch Later'
1552 _PERSONAL_FEED
= True
1554 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
1555 IE_DESC
= 'Youtube watch history, "ythistory" keyword (requires authentication)'
1556 _VALID_URL
= 'https?://www\.youtube\.com/feed/history|:ythistory'
1557 _FEED_NAME
= 'history'
1558 _PERSONAL_FEED
= True
1559 _PLAYLIST_TITLE
= 'Youtube Watch History'
1561 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
1562 IE_NAME
= 'youtube:favorites'
1563 IE_DESC
= 'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1564 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1565 _LOGIN_REQUIRED
= True
1567 def _real_extract(self
, url
):
1568 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1569 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, 'favourites playlist id')
1570 return self
.url_result(playlist_id
, 'YoutubePlaylist')
1573 class YoutubeSubscriptionsIE(YoutubePlaylistIE
):
1574 IE_NAME
= 'youtube:subscriptions'
1575 IE_DESC
= 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1576 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1579 def _real_extract(self
, url
):
1580 title
= 'Youtube Subscriptions'
1581 page
= self
._download
_webpage
('https://www.youtube.com/feed/subscriptions', title
)
1583 # The extraction process is the same as for playlists, but the regex
1584 # for the video ids doesn't contain an index
1586 more_widget_html
= content_html
= page
1588 for page_num
in itertools
.count(1):
1589 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
1590 new_ids
= orderedSet(matches
)
1593 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1597 more
= self
._download
_json
(
1598 'https://youtube.com/%s' % mobj
.group('more'), title
,
1599 'Downloading page #%s' % page_num
,
1600 transform_source
=uppercase_escape
)
1601 content_html
= more
['content_html']
1602 more_widget_html
= more
['load_more_widget_html']
1605 '_type': 'playlist',
1607 'entries': self
._ids
_to
_results
(ids
),
1611 class YoutubeTruncatedURLIE(InfoExtractor
):
1612 IE_NAME
= 'youtube:truncated_url'
1613 IE_DESC
= False # Do not list
1614 _VALID_URL
= r
'''(?x)
1615 (?:https?://)?[^/]+/watch\?(?:
1617 annotation_id=annotation_[^&]+
1619 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1623 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1624 'only_matching': True,
1626 'url': 'http://www.youtube.com/watch?',
1627 'only_matching': True,
1630 def _real_extract(self
, url
):
1631 raise ExtractorError(
1632 'Did you forget to quote the URL? Remember that & is a meta '
1633 'character in most shells, so you want to put the URL in quotes, '
1635 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1636 ' or simply youtube-dl BaW_jenozKc .',