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},
279 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
280 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
281 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
284 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
285 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
286 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
287 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
288 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
289 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
290 '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40, 'container': 'webm', 'vcodec': 'VP9'},
291 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
293 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
298 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
299 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
302 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
303 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
306 '_rtmp
': {'protocol': 'rtmp'},
312 'url
': 'http
://www
.youtube
.com
/watch?v
=BaW_jenozKc
',
316 'title
': 'youtube
-dl test video
"\'/\\ä↭𝕐',
317 'uploader': 'Philipp Hagemeister',
318 'uploader_id': 'phihag',
319 'upload_date': '20121002',
320 '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 .',
321 'categories
': ['Science
& Technology
'],
323 'dislike_count
': int,
327 'url
': 'http
://www
.youtube
.com
/watch?v
=UxxajLWwzqY
',
328 'note
': 'Test generic use_cipher_signature
video (#897)',
332 'upload_date': '20120506',
333 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
334 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
335 'uploader': 'Icona Pop',
336 'uploader_id': 'IconaPop',
340 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
341 'note': 'Test VEVO video with age protection (#956)',
345 'upload_date': '20130703',
346 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
347 'description': 'md5:64249768eec3bc4276236606ea996373',
348 'uploader': 'justintimberlakeVEVO',
349 'uploader_id': 'justintimberlakeVEVO',
353 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
354 'note': 'Embed-only video (#1746)',
358 'upload_date': '20120608',
359 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
360 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
361 'uploader': 'SET India',
362 'uploader_id': 'setindia'
366 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
367 'note': '256k DASH audio (format 141) via DASH manifest',
371 'upload_date': '20121002',
372 'uploader_id': '8KVIDEO',
374 'uploader': '8KVIDEO',
375 'title': 'UHDTV TEST 8K VIDEO.mp4'
378 'youtube_include_dash_manifest': True,
382 # DASH manifest with encrypted signature
384 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
388 'title': 'Afrojack - The Spark ft. Spree Wilson',
389 'description': 'md5:9717375db5a9a3992be4668bbf3bc0a8',
390 'uploader': 'AfrojackVEVO',
391 'uploader_id': 'AfrojackVEVO',
392 'upload_date': '20131011',
395 'youtube_include_dash_manifest': True,
401 def __init__(self
, *args
, **kwargs
):
402 super(YoutubeIE
, self
).__init
__(*args
, **kwargs
)
403 self
._player
_cache
= {}
405 def report_video_info_webpage_download(self
, video_id
):
406 """Report attempt to download video info webpage."""
407 self
.to_screen('%s: Downloading video info webpage' % video_id
)
409 def report_information_extraction(self
, video_id
):
410 """Report attempt to extract video information."""
411 self
.to_screen('%s: Extracting video information' % video_id
)
413 def report_unavailable_format(self
, video_id
, format
):
414 """Report extracted video URL."""
415 self
.to_screen('%s: Format %s not available' % (video_id
, format
))
417 def report_rtmp_download(self
):
418 """Indicate the download will use the RTMP protocol."""
419 self
.to_screen('RTMP download detected')
421 def _signature_cache_id(self
, example_sig
):
422 """ Return a string representation of a signature """
423 return '.'.join(compat_str(len(part
)) for part
in example_sig
.split('.'))
425 def _extract_signature_function(self
, video_id
, player_url
, example_sig
):
427 r
'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
430 raise ExtractorError('Cannot identify player %r' % player_url
)
431 player_type
= id_m
.group('ext')
432 player_id
= id_m
.group('id')
434 # Read from filesystem cache
435 func_id
= '%s_%s_%s' % (
436 player_type
, player_id
, self
._signature
_cache
_id
(example_sig
))
437 assert os
.path
.basename(func_id
) == func_id
439 cache_spec
= self
._downloader
.cache
.load('youtube-sigfuncs', func_id
)
440 if cache_spec
is not None:
441 return lambda s
: ''.join(s
[i
] for i
in cache_spec
)
443 if player_type
== 'js':
444 code
= self
._download
_webpage
(
445 player_url
, video_id
,
446 note
='Downloading %s player %s' % (player_type
, player_id
),
447 errnote
='Download of %s failed' % player_url
)
448 res
= self
._parse
_sig
_js
(code
)
449 elif player_type
== 'swf':
450 urlh
= self
._request
_webpage
(
451 player_url
, video_id
,
452 note
='Downloading %s player %s' % (player_type
, player_id
),
453 errnote
='Download of %s failed' % player_url
)
455 res
= self
._parse
_sig
_swf
(code
)
457 assert False, 'Invalid player type %r' % player_type
459 if cache_spec
is None:
460 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
461 cache_res
= res(test_string
)
462 cache_spec
= [ord(c
) for c
in cache_res
]
464 self
._downloader
.cache
.store('youtube-sigfuncs', func_id
, cache_spec
)
467 def _print_sig_code(self
, func
, example_sig
):
468 def gen_sig_code(idxs
):
469 def _genslice(start
, end
, step
):
470 starts
= '' if start
== 0 else str(start
)
471 ends
= (':%d' % (end
+step
)) if end
+ step
>= 0 else ':'
472 steps
= '' if step
== 1 else (':%d' % step
)
473 return 's[%s%s%s]' % (starts
, ends
, steps
)
476 start
= '(Never used)' # Quelch pyflakes warnings - start will be
477 # set as soon as step is set
478 for i
, prev
in zip(idxs
[1:], idxs
[:-1]):
482 yield _genslice(start
, prev
, step
)
485 if i
- prev
in [-1, 1]:
494 yield _genslice(start
, i
, step
)
496 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
497 cache_res
= func(test_string
)
498 cache_spec
= [ord(c
) for c
in cache_res
]
499 expr_code
= ' + '.join(gen_sig_code(cache_spec
))
500 signature_id_tuple
= '(%s)' % (
501 ', '.join(compat_str(len(p
)) for p
in example_sig
.split('.')))
502 code
= ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
503 ' return %s\n') % (signature_id_tuple
, expr_code
)
504 self
.to_screen('Extracted signature function:\n' + code
)
506 def _parse_sig_js(self
, jscode
):
507 funcname
= self
._search
_regex
(
508 r
'signature=([$a-zA-Z]+)', jscode
,
509 'Initial JS player signature function name')
511 jsi
= JSInterpreter(jscode
)
512 initial_function
= jsi
.extract_function(funcname
)
513 return lambda s
: initial_function([s
])
515 def _parse_sig_swf(self
, file_contents
):
516 swfi
= SWFInterpreter(file_contents
)
517 TARGET_CLASSNAME
= 'SignatureDecipher'
518 searched_class
= swfi
.extract_class(TARGET_CLASSNAME
)
519 initial_function
= swfi
.extract_function(searched_class
, 'decipher')
520 return lambda s
: initial_function([s
])
522 def _decrypt_signature(self
, s
, video_id
, player_url
, age_gate
=False):
523 """Turn the encrypted s field into a working signature"""
525 if player_url
is None:
526 raise ExtractorError('Cannot decrypt signature without player_url')
528 if player_url
.startswith('//'):
529 player_url
= 'https:' + player_url
531 player_id
= (player_url
, self
._signature
_cache
_id
(s
))
532 if player_id
not in self
._player
_cache
:
533 func
= self
._extract
_signature
_function
(
534 video_id
, player_url
, s
536 self
._player
_cache
[player_id
] = func
537 func
= self
._player
_cache
[player_id
]
538 if self
._downloader
.params
.get('youtube_print_sig_code'):
539 self
._print
_sig
_code
(func
, s
)
541 except Exception as e
:
542 tb
= traceback
.format_exc()
543 raise ExtractorError(
544 'Signature extraction failed: ' + tb
, cause
=e
)
546 def _get_available_subtitles(self
, video_id
, webpage
):
548 sub_list
= self
._download
_webpage
(
549 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
,
550 video_id
, note
=False)
551 except ExtractorError
as err
:
552 self
._downloader
.report_warning('unable to download video subtitles: %s' % compat_str(err
))
554 lang_list
= re
.findall(r
'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list
)
559 if lang
in sub_lang_list
:
561 params
= compat_urllib_parse
.urlencode({
564 'fmt': self
._downloader
.params
.get('subtitlesformat', 'srt'),
565 'name': unescapeHTML(l
[0]).encode('utf-8'),
567 url
= 'https://www.youtube.com/api/timedtext?' + params
568 sub_lang_list
[lang
] = url
569 if not sub_lang_list
:
570 self
._downloader
.report_warning('video doesn\'t have subtitles')
574 def _get_available_automatic_caption(self
, video_id
, webpage
):
575 """We need the webpage for getting the captions url, pass it as an
576 argument to speed up the process."""
577 sub_format
= self
._downloader
.params
.get('subtitlesformat', 'srt')
578 self
.to_screen('%s: Looking for automatic captions' % video_id
)
579 mobj
= re
.search(r
';ytplayer.config = ({.*?});', webpage
)
580 err_msg
= 'Couldn\'t find automatic captions for %s' % video_id
582 self
._downloader
.report_warning(err_msg
)
584 player_config
= json
.loads(mobj
.group(1))
586 args
= player_config
[u
'args']
587 caption_url
= args
[u
'ttsurl']
588 timestamp
= args
[u
'timestamp']
589 # We get the available subtitles
590 list_params
= compat_urllib_parse
.urlencode({
595 list_url
= caption_url
+ '&' + list_params
596 caption_list
= self
._download
_xml
(list_url
, video_id
)
597 original_lang_node
= caption_list
.find('track')
598 if original_lang_node
is None or original_lang_node
.attrib
.get('kind') != 'asr' :
599 self
._downloader
.report_warning('Video doesn\'t have automatic captions')
601 original_lang
= original_lang_node
.attrib
['lang_code']
604 for lang_node
in caption_list
.findall('target'):
605 sub_lang
= lang_node
.attrib
['lang_code']
606 params
= compat_urllib_parse
.urlencode({
607 'lang': original_lang
,
613 sub_lang_list
[sub_lang
] = caption_url
+ '&' + params
615 # An extractor error can be raise by the download process if there are
616 # no automatic captions but there are subtitles
617 except (KeyError, ExtractorError
):
618 self
._downloader
.report_warning(err_msg
)
622 def extract_id(cls
, url
):
623 mobj
= re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
)
625 raise ExtractorError('Invalid URL: %s' % url
)
626 video_id
= mobj
.group(2)
629 def _extract_from_m3u8(self
, manifest_url
, video_id
):
631 def _get_urls(_manifest
):
632 lines
= _manifest
.split('\n')
633 urls
= filter(lambda l
: l
and not l
.startswith('#'),
636 manifest
= self
._download
_webpage
(manifest_url
, video_id
, 'Downloading formats manifest')
637 formats_urls
= _get_urls(manifest
)
638 for format_url
in formats_urls
:
639 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
640 url_map
[itag
] = format_url
643 def _extract_annotations(self
, video_id
):
644 url
= 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
645 return self
._download
_webpage
(url
, video_id
, note
='Searching for annotations.', errnote
='Unable to download video annotations.')
647 def _real_extract(self
, url
):
649 'http' if self
._downloader
.params
.get('prefer_insecure', False)
652 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
653 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
655 url
= proto
+ '://www.youtube.com/' + compat_urllib_parse
.unquote(mobj
.group(1)).lstrip('/')
656 video_id
= self
.extract_id(url
)
659 url
= proto
+ '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
661 c
for c
in self
._downloader
.cookiejar
662 if c
.domain
== '.youtube.com' and c
.name
== 'PREF']
663 for pc
in pref_cookies
:
664 if 'hl=' in pc
.value
:
665 pc
.value
= re
.sub(r
'hl=[^&]+', 'hl=en', pc
.value
)
670 video_webpage
= self
._download
_webpage
(url
, video_id
)
672 # Attempt to extract SWF player URL
673 mobj
= re
.search(r
'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
675 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
680 self
.report_video_info_webpage_download(video_id
)
681 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
682 self
.report_age_confirmation()
684 # We simulate the access to the video from www.youtube.com/v/{video_id}
685 # this can be viewed without login into Youtube
686 data
= compat_urllib_parse
.urlencode({
687 'video_id': video_id
,
688 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
689 'sts': self
._search
_regex
(
690 r
'"sts"\s*:\s*(\d+)', video_webpage
, 'sts'),
692 video_info_url
= proto
+ '://www.youtube.com/get_video_info?' + data
693 video_info_webpage
= self
._download
_webpage
(video_info_url
, video_id
,
695 errnote
='unable to download video info webpage')
696 video_info
= compat_parse_qs(video_info_webpage
)
699 for el_type
in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
700 video_info_url
= (proto
+ '://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
701 % (video_id
, el_type
))
702 video_info_webpage
= self
._download
_webpage
(video_info_url
, video_id
,
704 errnote
='unable to download video info webpage')
705 video_info
= compat_parse_qs(video_info_webpage
)
706 if 'token' in video_info
:
708 if 'token' not in video_info
:
709 if 'reason' in video_info
:
710 raise ExtractorError(
711 'YouTube said: %s' % video_info
['reason'][0],
712 expected
=True, video_id
=video_id
)
714 raise ExtractorError(
715 '"token" parameter not in video info for unknown reason',
718 if 'view_count' in video_info
:
719 view_count
= int(video_info
['view_count'][0])
723 # Check for "rental" videos
724 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
725 raise ExtractorError('"rental" videos not supported')
727 # Start extracting information
728 self
.report_information_extraction(video_id
)
731 if 'author' not in video_info
:
732 raise ExtractorError('Unable to extract uploader name')
733 video_uploader
= compat_urllib_parse
.unquote_plus(video_info
['author'][0])
736 video_uploader_id
= None
737 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
739 video_uploader_id
= mobj
.group(1)
741 self
._downloader
.report_warning('unable to extract uploader nickname')
744 if 'title' in video_info
:
745 video_title
= video_info
['title'][0]
747 self
._downloader
.report_warning('Unable to extract video title')
751 # We try first to get a high quality image:
752 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
753 video_webpage
, re
.DOTALL
)
754 if m_thumb
is not None:
755 video_thumbnail
= m_thumb
.group(1)
756 elif 'thumbnail_url' not in video_info
:
757 self
._downloader
.report_warning('unable to extract video thumbnail')
758 video_thumbnail
= None
759 else: # don't panic if we can't find it
760 video_thumbnail
= compat_urllib_parse
.unquote_plus(video_info
['thumbnail_url'][0])
764 mobj
= re
.search(r
'(?s)id="eow-date.*?>(.*?)</span>', video_webpage
)
767 r
'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
770 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
771 upload_date
= unified_strdate(upload_date
)
773 m_cat_container
= self
._search
_regex
(
774 r
'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
775 video_webpage
, 'categories', fatal
=False)
777 category
= self
._html
_search
_regex
(
778 r
'(?s)<a[^<]+>(.*?)</a>', m_cat_container
, 'category',
780 video_categories
= None if category
is None else [category
]
782 video_categories
= None
785 video_description
= get_element_by_id("eow-description", video_webpage
)
786 if video_description
:
787 video_description
= re
.sub(r
'''(?x)
789 (?:[a-zA-Z-]+="[^"]+"\s+)*?
791 (?:[a-zA-Z-]+="[^"]+"\s+)*?
792 class="yt-uix-redirect-link"\s*>
795 ''', r
'\1', video_description
)
796 video_description
= clean_html(video_description
)
798 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
800 video_description
= unescapeHTML(fd_mobj
.group(1))
802 video_description
= ''
804 def _extract_count(count_name
):
805 count
= self
._search
_regex
(
806 r
'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re
.escape(count_name
),
807 video_webpage
, count_name
, default
=None)
808 if count
is not None:
809 return int(count
.replace(',', ''))
811 like_count
= _extract_count('like')
812 dislike_count
= _extract_count('dislike')
815 video_subtitles
= self
.extract_subtitles(video_id
, video_webpage
)
817 if self
._downloader
.params
.get('listsubtitles', False):
818 self
._list
_available
_subtitles
(video_id
, video_webpage
)
821 if 'length_seconds' not in video_info
:
822 self
._downloader
.report_warning('unable to extract video duration')
823 video_duration
= None
825 video_duration
= int(compat_urllib_parse
.unquote_plus(video_info
['length_seconds'][0]))
828 video_annotations
= None
829 if self
._downloader
.params
.get('writeannotations', False):
830 video_annotations
= self
._extract
_annotations
(video_id
)
832 # Decide which formats to download
834 mobj
= re
.search(r
';ytplayer\.config\s*=\s*({.*?});', video_webpage
)
836 raise ValueError('Could not find vevo ID')
837 json_code
= uppercase_escape(mobj
.group(1))
838 ytplayer_config
= json
.loads(json_code
)
839 args
= ytplayer_config
['args']
840 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
841 # this signatures are encrypted
842 if 'url_encoded_fmt_stream_map' not in args
:
843 raise ValueError('No stream_map present') # caught below
844 re_signature
= re
.compile(r
'[&,]s=')
845 m_s
= re_signature
.search(args
['url_encoded_fmt_stream_map'])
847 self
.to_screen('%s: Encrypted signatures detected.' % video_id
)
848 video_info
['url_encoded_fmt_stream_map'] = [args
['url_encoded_fmt_stream_map']]
849 m_s
= re_signature
.search(args
.get('adaptive_fmts', ''))
851 if 'adaptive_fmts' in video_info
:
852 video_info
['adaptive_fmts'][0] += ',' + args
['adaptive_fmts']
854 video_info
['adaptive_fmts'] = [args
['adaptive_fmts']]
858 def _map_to_format_list(urlmap
):
860 for itag
, video_real_url
in urlmap
.items():
863 'url': video_real_url
,
864 'player_url': player_url
,
866 if itag
in self
._formats
:
867 dct
.update(self
._formats
[itag
])
871 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
872 self
.report_rtmp_download()
874 'format_id': '_rtmp',
876 'url': video_info
['conn'][0],
877 'player_url': player_url
,
879 elif len(video_info
.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info
.get('adaptive_fmts', [])) >= 1:
880 encoded_url_map
= video_info
.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info
.get('adaptive_fmts',[''])[0]
881 if 'rtmpe%3Dyes' in encoded_url_map
:
882 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
884 for url_data_str
in encoded_url_map
.split(','):
885 url_data
= compat_parse_qs(url_data_str
)
886 if 'itag' not in url_data
or 'url' not in url_data
:
888 format_id
= url_data
['itag'][0]
889 url
= url_data
['url'][0]
891 if 'sig' in url_data
:
892 url
+= '&signature=' + url_data
['sig'][0]
893 elif 's' in url_data
:
894 encrypted_sig
= url_data
['s'][0]
897 jsplayer_url_json
= self
._search
_regex
(
898 r
'"assets":.+?"js":\s*("[^"]+")',
899 video_webpage
, 'JS player URL')
900 player_url
= json
.loads(jsplayer_url_json
)
901 if player_url
is None:
902 player_url_json
= self
._search
_regex
(
903 r
'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
904 video_webpage
, 'age gate player URL')
905 player_url
= json
.loads(player_url_json
)
907 if self
._downloader
.params
.get('verbose'):
908 if player_url
is None:
909 player_version
= 'unknown'
910 player_desc
= 'unknown'
912 if player_url
.endswith('swf'):
913 player_version
= self
._search
_regex
(
914 r
'-(.+?)(?:/watch_as3)?\.swf$', player_url
,
915 'flash player', fatal
=False)
916 player_desc
= 'flash player %s' % player_version
918 player_version
= self
._search
_regex
(
919 r
'html5player-([^/]+?)(?:/html5player)?\.js',
921 'html5 player', fatal
=False)
922 player_desc
= 'html5 player %s' % player_version
924 parts_sizes
= self
._signature
_cache
_id
(encrypted_sig
)
925 self
.to_screen('{%s} signature length %s, %s' %
926 (format_id
, parts_sizes
, player_desc
))
928 signature
= self
._decrypt
_signature
(
929 encrypted_sig
, video_id
, player_url
, age_gate
)
930 url
+= '&signature=' + signature
931 if 'ratebypass' not in url
:
932 url
+= '&ratebypass=yes'
933 url_map
[format_id
] = url
934 formats
= _map_to_format_list(url_map
)
935 elif video_info
.get('hlsvp'):
936 manifest_url
= video_info
['hlsvp'][0]
937 url_map
= self
._extract
_from
_m
3u8(manifest_url
, video_id
)
938 formats
= _map_to_format_list(url_map
)
940 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
942 # Look for the DASH manifest
943 if self
._downloader
.params
.get('youtube_include_dash_manifest', True):
945 # The DASH manifest used needs to be the one from the original video_webpage.
946 # The one found in get_video_info seems to be using different signatures.
947 # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
948 # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
949 # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
951 dash_manifest_url
= video_info
.get('dashmpd')[0]
953 dash_manifest_url
= ytplayer_config
['args']['dashmpd']
954 def decrypt_sig(mobj
):
956 dec_s
= self
._decrypt
_signature
(s
, video_id
, player_url
, age_gate
)
957 return '/signature/%s' % dec_s
958 dash_manifest_url
= re
.sub(r
'/s/([\w\.]+)', decrypt_sig
, dash_manifest_url
)
959 dash_doc
= self
._download
_xml
(
960 dash_manifest_url
, video_id
,
961 note
='Downloading DASH manifest',
962 errnote
='Could not download DASH manifest')
963 for r
in dash_doc
.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
964 url_el
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
967 format_id
= r
.attrib
['id']
968 video_url
= url_el
.text
969 filesize
= int_or_none(url_el
.attrib
.get('{http://youtube.com/yt/2012/10/10}contentLength'))
971 'format_id': format_id
,
973 'width': int_or_none(r
.attrib
.get('width')),
974 'tbr': int_or_none(r
.attrib
.get('bandwidth'), 1000),
975 'asr': int_or_none(r
.attrib
.get('audioSamplingRate')),
976 'filesize': filesize
,
979 existing_format
= next(
981 if fo
['format_id'] == format_id
)
982 except StopIteration:
983 f
.update(self
._formats
.get(format_id
, {}))
986 existing_format
.update(f
)
988 except (ExtractorError
, KeyError) as e
:
989 self
.report_warning('Skipping DASH manifest: %s' % e
, video_id
)
991 self
._sort
_formats
(formats
)
995 'uploader': video_uploader
,
996 'uploader_id': video_uploader_id
,
997 'upload_date': upload_date
,
998 'title': video_title
,
999 'thumbnail': video_thumbnail
,
1000 'description': video_description
,
1001 'categories': video_categories
,
1002 'subtitles': video_subtitles
,
1003 'duration': video_duration
,
1004 'age_limit': 18 if age_gate
else 0,
1005 'annotations': video_annotations
,
1006 'webpage_url': proto
+ '://www.youtube.com/watch?v=%s' % video_id
,
1007 'view_count': view_count
,
1008 'like_count': like_count
,
1009 'dislike_count': dislike_count
,
1013 class YoutubePlaylistIE(YoutubeBaseInfoExtractor
):
1014 IE_DESC
= 'YouTube.com playlists'
1015 _VALID_URL
= r
"""(?x)(?:
1020 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1021 \? (?:.*?&)*? (?:p|a|list)=
1025 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1026 # Top tracks, they can also include dots
1031 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1033 _TEMPLATE_URL
= 'https://www.youtube.com/playlist?list=%s'
1034 _MORE_PAGES_INDICATOR
= r
'data-link-type="next"'
1035 _VIDEO_RE
= r
'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1036 IE_NAME
= 'youtube:playlist'
1038 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1040 'title': 'ytdl test PL',
1042 'playlist_count': 3,
1044 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1046 'title': 'YDL_Empty_List',
1048 'playlist_count': 0,
1050 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1051 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1053 'title': '29C3: Not my department',
1055 'playlist_count': 95,
1057 'note': 'issue #673',
1058 'url': 'PLBB231211A4F62143',
1060 'title': '[OLD]Team Fortress 2 (Class-based LP)',
1062 'playlist_mincount': 26,
1064 'note': 'Large playlist',
1065 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1067 'title': 'Uploads from Cauchemar',
1069 'playlist_mincount': 799,
1071 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1073 'title': 'YDL_safe_search',
1075 'playlist_count': 2,
1078 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1079 'playlist_count': 4,
1084 'note': 'Embedded SWF player',
1085 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1086 'playlist_count': 4,
1092 def _real_initialize(self
):
1095 def _ids_to_results(self
, ids
):
1097 self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
1100 def _extract_mix(self
, playlist_id
):
1101 # The mixes are generated from a a single video
1102 # the id of the playlist is just 'RD' + video_id
1103 url
= 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id
[-11:], playlist_id
)
1104 webpage
= self
._download
_webpage
(
1105 url
, playlist_id
, 'Downloading Youtube mix')
1106 search_title
= lambda class_name
: get_element_by_attribute('class', class_name
, webpage
)
1108 search_title('playlist-title') or
1109 search_title('title long-title') or
1110 search_title('title'))
1111 title
= clean_html(title_span
)
1112 ids
= orderedSet(re
.findall(
1113 r
'''(?xs)data-video-username=".*?".*?
1114 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re
.escape(playlist_id
),
1116 url_results
= self
._ids
_to
_results
(ids
)
1118 return self
.playlist_result(url_results
, playlist_id
, title
)
1120 def _real_extract(self
, url
):
1121 # Extract playlist id
1122 mobj
= re
.match(self
._VALID
_URL
, url
)
1124 raise ExtractorError('Invalid URL: %s' % url
)
1125 playlist_id
= mobj
.group(1) or mobj
.group(2)
1127 # Check if it's a video-specific URL
1128 query_dict
= compat_urlparse
.parse_qs(compat_urlparse
.urlparse(url
).query
)
1129 if 'v' in query_dict
:
1130 video_id
= query_dict
['v'][0]
1131 if self
._downloader
.params
.get('noplaylist'):
1132 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1133 return self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1135 self
.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id
, video_id
))
1137 if playlist_id
.startswith('RD'):
1138 # Mixes require a custom extraction process
1139 return self
._extract
_mix
(playlist_id
)
1140 if playlist_id
.startswith('TL'):
1141 raise ExtractorError('For downloading YouTube.com top lists, use '
1142 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected
=True)
1144 url
= self
._TEMPLATE
_URL
% playlist_id
1145 page
= self
._download
_webpage
(url
, playlist_id
)
1146 more_widget_html
= content_html
= page
1148 # Check if the playlist exists or is private
1149 if re
.search(r
'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page
) is not None:
1150 raise ExtractorError(
1151 'The playlist doesn\'t exist or is private, use --username or '
1152 '--netrc to access it.',
1155 # Extract the video ids from the playlist pages
1158 for page_num
in itertools
.count(1):
1159 matches
= re
.finditer(self
._VIDEO
_RE
, content_html
)
1160 # We remove the duplicates and the link with index 0
1161 # (it's not the first video of the playlist)
1162 new_ids
= orderedSet(m
.group('id') for m
in matches
if m
.group('index') != '0')
1165 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1169 more
= self
._download
_json
(
1170 'https://youtube.com/%s' % mobj
.group('more'), playlist_id
,
1171 'Downloading page #%s' % page_num
,
1172 transform_source
=uppercase_escape
)
1173 content_html
= more
['content_html']
1174 more_widget_html
= more
['load_more_widget_html']
1176 playlist_title
= self
._html
_search
_regex
(
1177 r
'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1180 url_results
= self
._ids
_to
_results
(ids
)
1181 return self
.playlist_result(url_results
, playlist_id
, playlist_title
)
1184 class YoutubeTopListIE(YoutubePlaylistIE
):
1185 IE_NAME
= 'youtube:toplist'
1186 IE_DESC
= ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1187 ' (Example: "yttoplist:music:Top Tracks")')
1188 _VALID_URL
= r
'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1190 'url': 'yttoplist:music:Trending',
1191 'playlist_mincount': 5,
1192 'skip': 'Only works for logged-in users',
1195 def _real_extract(self
, url
):
1196 mobj
= re
.match(self
._VALID
_URL
, url
)
1197 channel
= mobj
.group('chann')
1198 title
= mobj
.group('title')
1199 query
= compat_urllib_parse
.urlencode({'title': title}
)
1200 channel_page
= self
._download
_webpage
(
1201 'https://www.youtube.com/%s' % channel
, title
)
1202 link
= self
._html
_search
_regex
(
1204 <a\s+href="([^"]+)".*?>\s*
1205 <span\s+class="branded-page-module-title-text">\s*
1206 <span[^>]*>.*?%s.*?</span>''' % re
.escape(query
),
1207 channel_page
, 'list')
1208 url
= compat_urlparse
.urljoin('https://www.youtube.com/', link
)
1210 video_re
= r
'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1212 # sometimes the webpage doesn't contain the videos
1213 # retry until we get them
1214 for i
in itertools
.count(0):
1215 msg
= 'Downloading Youtube mix'
1217 msg
+= ', retry #%d' % i
1219 webpage
= self
._download
_webpage
(url
, title
, msg
)
1220 ids
= orderedSet(re
.findall(video_re
, webpage
))
1223 url_results
= self
._ids
_to
_results
(ids
)
1224 return self
.playlist_result(url_results
, playlist_title
=title
)
1227 class YoutubeChannelIE(InfoExtractor
):
1228 IE_DESC
= 'YouTube.com channels'
1229 _VALID_URL
= r
"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
1230 _MORE_PAGES_INDICATOR
= 'yt-uix-load-more'
1231 _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'
1232 IE_NAME
= 'youtube:channel'
1234 'note': 'paginated channel',
1235 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1236 'playlist_mincount': 91,
1239 def extract_videos_from_page(self
, page
):
1241 for mobj
in re
.finditer(r
'href="/watch\?v=([0-9A-Za-z_-]+)&?', page
):
1242 if mobj
.group(1) not in ids_in_page
:
1243 ids_in_page
.append(mobj
.group(1))
1246 def _real_extract(self
, url
):
1247 # Extract channel id
1248 mobj
= re
.match(self
._VALID
_URL
, url
)
1250 raise ExtractorError('Invalid URL: %s' % url
)
1252 # Download channel page
1253 channel_id
= mobj
.group(1)
1255 url
= 'https://www.youtube.com/channel/%s/videos' % channel_id
1256 channel_page
= self
._download
_webpage
(url
, channel_id
)
1257 autogenerated
= re
.search(r
'''(?x)
1259 channel-header-autogenerated-label|
1260 yt-channel-title-autogenerated
1261 )[^"]*"''', channel_page
) is not None
1264 # The videos are contained in a single page
1265 # the ajax pages can't be used, they are empty
1266 video_ids
= self
.extract_videos_from_page(channel_page
)
1268 # Download all channel pages using the json-based channel_ajax query
1269 for pagenum
in itertools
.count(1):
1270 url
= self
._MORE
_PAGES
_URL
% (pagenum
, channel_id
)
1271 page
= self
._download
_json
(
1272 url
, channel_id
, note
='Downloading page #%s' % pagenum
,
1273 transform_source
=uppercase_escape
)
1275 ids_in_page
= self
.extract_videos_from_page(page
['content_html'])
1276 video_ids
.extend(ids_in_page
)
1278 if self
._MORE
_PAGES
_INDICATOR
not in page
['load_more_widget_html']:
1281 self
._downloader
.to_screen('[youtube] Channel %s: Found %i videos' % (channel_id
, len(video_ids
)))
1283 url_entries
= [self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1284 for video_id
in video_ids
]
1285 return self
.playlist_result(url_entries
, channel_id
)
1288 class YoutubeUserIE(InfoExtractor
):
1289 IE_DESC
= 'YouTube.com user videos (URL or "ytuser" keyword)'
1290 _VALID_URL
= r
'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
1291 _TEMPLATE_URL
= 'https://gdata.youtube.com/feeds/api/users/%s'
1292 _GDATA_PAGE_SIZE
= 50
1293 _GDATA_URL
= 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1294 IE_NAME
= 'youtube:user'
1297 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1298 'playlist_mincount': 320,
1300 'title': 'TheLinuxFoundation',
1303 'url': 'ytuser:phihag',
1304 'only_matching': True,
1308 def suitable(cls
, url
):
1309 # Don't return True if the url can be extracted with other youtube
1310 # extractor, the regex would is too permissive and it would match.
1311 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1312 if any(ie
.suitable(url
) for ie
in other_ies
): return False
1313 else: return super(YoutubeUserIE
, cls
).suitable(url
)
1315 def _real_extract(self
, url
):
1317 mobj
= re
.match(self
._VALID
_URL
, url
)
1319 raise ExtractorError('Invalid URL: %s' % url
)
1321 username
= mobj
.group(1)
1323 # Download video ids using YouTube Data API. Result size per
1324 # query is limited (currently to 50 videos) so we need to query
1325 # page by page until there are no video ids - it means we got
1328 def download_page(pagenum
):
1329 start_index
= pagenum
* self
._GDATA
_PAGE
_SIZE
+ 1
1331 gdata_url
= self
._GDATA
_URL
% (username
, self
._GDATA
_PAGE
_SIZE
, start_index
)
1332 page
= self
._download
_webpage
(
1333 gdata_url
, username
,
1334 'Downloading video ids from %d to %d' % (
1335 start_index
, start_index
+ self
._GDATA
_PAGE
_SIZE
))
1338 response
= json
.loads(page
)
1339 except ValueError as err
:
1340 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err
))
1341 if 'entry' not in response
['feed']:
1344 # Extract video identifiers
1345 entries
= response
['feed']['entry']
1346 for entry
in entries
:
1347 title
= entry
['title']['$t']
1348 video_id
= entry
['id']['$t'].split('/')[-1]
1352 'ie_key': 'Youtube',
1356 url_results
= OnDemandPagedList(download_page
, self
._GDATA
_PAGE
_SIZE
)
1358 return self
.playlist_result(url_results
, playlist_title
=username
)
1361 class YoutubeSearchIE(SearchInfoExtractor
):
1362 IE_DESC
= 'YouTube.com searches'
1363 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1365 IE_NAME
= 'youtube:search'
1366 _SEARCH_KEY
= 'ytsearch'
1368 def _get_n_results(self
, query
, n
):
1369 """Get a specified number of results for a query"""
1376 while (PAGE_SIZE
* pagenum
) < limit
:
1377 result_url
= self
._API
_URL
% (
1378 compat_urllib_parse
.quote_plus(query
.encode('utf-8')),
1379 (PAGE_SIZE
* pagenum
) + 1)
1380 data_json
= self
._download
_webpage
(
1381 result_url
, video_id
='query "%s"' % query
,
1382 note
='Downloading page %s' % (pagenum
+ 1),
1383 errnote
='Unable to download API page')
1384 data
= json
.loads(data_json
)
1385 api_response
= data
['data']
1387 if 'items' not in api_response
:
1388 raise ExtractorError(
1389 '[youtube] No video results', expected
=True)
1391 new_ids
= list(video
['id'] for video
in api_response
['items'])
1392 video_ids
+= new_ids
1394 limit
= min(n
, api_response
['totalItems'])
1397 if len(video_ids
) > n
:
1398 video_ids
= video_ids
[:n
]
1399 videos
= [self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1400 for video_id
in video_ids
]
1401 return self
.playlist_result(videos
, query
)
1404 class YoutubeSearchDateIE(YoutubeSearchIE
):
1405 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1406 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1407 _SEARCH_KEY
= 'ytsearchdate'
1408 IE_DESC
= 'YouTube.com searches, newest videos first'
1411 class YoutubeSearchURLIE(InfoExtractor
):
1412 IE_DESC
= 'YouTube.com search URLs'
1413 IE_NAME
= 'youtube:search_url'
1414 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1416 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1417 'playlist_mincount': 5,
1419 'title': 'youtube-dl test video',
1423 def _real_extract(self
, url
):
1424 mobj
= re
.match(self
._VALID
_URL
, url
)
1425 query
= compat_urllib_parse
.unquote_plus(mobj
.group('query'))
1427 webpage
= self
._download
_webpage
(url
, query
)
1428 result_code
= self
._search
_regex
(
1429 r
'(?s)<ol class="item-section"(.*?)</ol>', webpage
, 'result HTML')
1431 part_codes
= re
.findall(
1432 r
'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code
)
1434 for part_code
in part_codes
:
1435 part_title
= self
._html
_search
_regex
(
1436 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1437 part_url_snippet
= self
._html
_search
_regex
(
1438 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1439 part_url
= compat_urlparse
.urljoin(
1440 'https://www.youtube.com/', part_url_snippet
)
1444 'title': part_title
,
1448 '_type': 'playlist',
1454 class YoutubeShowIE(InfoExtractor
):
1455 IE_DESC
= 'YouTube.com (multi-season) shows'
1456 _VALID_URL
= r
'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1457 IE_NAME
= 'youtube:show'
1459 'url': 'http://www.youtube.com/show/airdisasters',
1460 'playlist_mincount': 3,
1462 'id': 'airdisasters',
1463 'title': 'Air Disasters',
1467 def _real_extract(self
, url
):
1468 mobj
= re
.match(self
._VALID
_URL
, url
)
1469 playlist_id
= mobj
.group('id')
1470 webpage
= self
._download
_webpage
(
1471 url
, playlist_id
, 'Downloading show webpage')
1472 # There's one playlist for each season of the show
1473 m_seasons
= list(re
.finditer(r
'href="(/playlist\?list=.*?)"', webpage
))
1474 self
.to_screen('%s: Found %s seasons' % (playlist_id
, len(m_seasons
)))
1477 'https://www.youtube.com' + season
.group(1), 'YoutubePlaylist')
1478 for season
in m_seasons
1480 title
= self
._og
_search
_title
(webpage
, fatal
=False)
1483 '_type': 'playlist',
1490 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1492 Base class for extractors that fetch info from
1493 http://www.youtube.com/feed_ajax
1494 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1496 _LOGIN_REQUIRED
= True
1497 # use action_load_personal_feed instead of action_load_system_feed
1498 _PERSONAL_FEED
= False
1501 def _FEED_TEMPLATE(self
):
1502 action
= 'action_load_system_feed'
1503 if self
._PERSONAL
_FEED
:
1504 action
= 'action_load_personal_feed'
1505 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action
, self
._FEED
_NAME
)
1509 return 'youtube:%s' % self
._FEED
_NAME
1511 def _real_initialize(self
):
1514 def _real_extract(self
, url
):
1517 for i
in itertools
.count(1):
1518 info
= self
._download
_json
(self
._FEED
_TEMPLATE
% paging
,
1519 '%s feed' % self
._FEED
_NAME
,
1520 'Downloading page %s' % i
)
1521 feed_html
= info
.get('feed_html') or info
.get('content_html')
1522 load_more_widget_html
= info
.get('load_more_widget_html') or feed_html
1523 m_ids
= re
.finditer(r
'"/watch\?v=(.*?)["&]', feed_html
)
1524 ids
= orderedSet(m
.group(1) for m
in m_ids
)
1525 feed_entries
.extend(
1526 self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1527 for video_id
in ids
)
1529 r
'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1530 load_more_widget_html
)
1533 paging
= mobj
.group('paging')
1534 return self
.playlist_result(feed_entries
, playlist_title
=self
._PLAYLIST
_TITLE
)
1536 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
1537 IE_DESC
= 'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1538 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1539 _FEED_NAME
= 'recommended'
1540 _PLAYLIST_TITLE
= 'Youtube Recommended videos'
1542 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor
):
1543 IE_DESC
= 'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1544 _VALID_URL
= r
'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1545 _FEED_NAME
= 'watch_later'
1546 _PLAYLIST_TITLE
= 'Youtube Watch Later'
1547 _PERSONAL_FEED
= True
1549 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
1550 IE_DESC
= 'Youtube watch history, "ythistory" keyword (requires authentication)'
1551 _VALID_URL
= 'https?://www\.youtube\.com/feed/history|:ythistory'
1552 _FEED_NAME
= 'history'
1553 _PERSONAL_FEED
= True
1554 _PLAYLIST_TITLE
= 'Youtube Watch History'
1556 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
1557 IE_NAME
= 'youtube:favorites'
1558 IE_DESC
= 'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1559 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1560 _LOGIN_REQUIRED
= True
1562 def _real_extract(self
, url
):
1563 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1564 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, 'favourites playlist id')
1565 return self
.url_result(playlist_id
, 'YoutubePlaylist')
1568 class YoutubeSubscriptionsIE(YoutubePlaylistIE
):
1569 IE_NAME
= 'youtube:subscriptions'
1570 IE_DESC
= 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1571 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1574 def _real_extract(self
, url
):
1575 title
= 'Youtube Subscriptions'
1576 page
= self
._download
_webpage
('https://www.youtube.com/feed/subscriptions', title
)
1578 # The extraction process is the same as for playlists, but the regex
1579 # for the video ids doesn't contain an index
1581 more_widget_html
= content_html
= page
1583 for page_num
in itertools
.count(1):
1584 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
1585 new_ids
= orderedSet(matches
)
1588 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1592 more
= self
._download
_json
(
1593 'https://youtube.com/%s' % mobj
.group('more'), title
,
1594 'Downloading page #%s' % page_num
,
1595 transform_source
=uppercase_escape
)
1596 content_html
= more
['content_html']
1597 more_widget_html
= more
['load_more_widget_html']
1600 '_type': 'playlist',
1602 'entries': self
._ids
_to
_results
(ids
),
1606 class YoutubeTruncatedURLIE(InfoExtractor
):
1607 IE_NAME
= 'youtube:truncated_url'
1608 IE_DESC
= False # Do not list
1609 _VALID_URL
= r
'''(?x)
1610 (?:https?://)?[^/]+/watch\?(?:
1612 annotation_id=annotation_[^&]+
1614 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1618 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1619 'only_matching': True,
1621 'url': 'http://www.youtube.com/watch?',
1622 'only_matching': True,
1625 def _real_extract(self
, url
):
1626 raise ExtractorError(
1627 'Did you forget to quote the URL? Remember that & is a meta '
1628 'character in most shells, so you want to put the URL in quotes, '
1630 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1631 ' or simply youtube-dl BaW_jenozKc .',