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 not self._set_language():
196 if not self._login():
201 class YoutubeIE(YoutubeBaseInfoExtractor, SubtitlesInfoExtractor):
202 IE_DESC = 'YouTube
.com
'
203 _VALID_URL = r"""(?x)^
205 (?:https?://|//) # http(s):// or protocol-independent URL
206 (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
207 (?:www\.)?deturl\.com/www\.youtube\.com/|
208 (?:www\.)?pwnyoutube\.com/|
209 (?:www\.)?yourepeat\.com/|
210 tube\.majestyc\.net/|
211 youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
212 (?:.*?\#/)? # handle anchor (#/) redirect urls
213 (?: # the various things that can precede the ID:
214 (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/
215 |(?: # or the v= param in all its forms
216 (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
217 (?:\?|\#!?) # the params delimiter ? or # or #!
218 (?:.*?&)? # any other preceding param (like /?s=tuff&v=xxxx)
222 |youtu\.be/ # just youtu.be/xxxx
223 |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
225 )? # all until now is optional -> you can pass the naked ID
226 ([0-9A-Za-z_-]{11}) # here is it! the YouTube video ID
227 (?!.*?&list=) # combined list/video URLs are handled by the playlist IE
228 (?(1).+)? # if we found the ID, everything can follow
230 _NEXT_URL_RE = r'[\?&]next_url
=([^
&]+)'
232 '5': {'ext': 'flv', 'width': 400, 'height': 240},
233 '6': {'ext': 'flv', 'width': 450, 'height': 270},
234 '13': {'ext': '3gp'},
235 '17': {'ext': '3gp', 'width': 176, 'height': 144},
236 '18': {'ext': 'mp4', 'width': 640, 'height': 360},
237 '22': {'ext': 'mp4', 'width': 1280, 'height': 720},
238 '34': {'ext': 'flv', 'width': 640, 'height': 360},
239 '35': {'ext': 'flv', 'width': 854, 'height': 480},
240 '36': {'ext': '3gp', 'width': 320, 'height': 240},
241 '37': {'ext': 'mp4', 'width': 1920, 'height': 1080},
242 '38': {'ext': 'mp4', 'width': 4096, 'height': 3072},
243 '43': {'ext': 'webm', 'width': 640, 'height': 360},
244 '44': {'ext': 'webm', 'width': 854, 'height': 480},
245 '45': {'ext': 'webm', 'width': 1280, 'height': 720},
246 '46': {'ext': 'webm', 'width': 1920, 'height': 1080},
250 '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'preference': -20},
251 '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'preference': -20},
252 '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'preference': -20},
253 '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'preference': -20},
254 '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'preference': -20},
255 '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'preference': -20},
256 '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'preference': -20},
258 # Apple HTTP Live Streaming
259 '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
260 '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'preference': -10},
261 '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'preference': -10},
262 '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'preference': -10},
263 '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'preference': -10},
264 '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'preference': -10},
265 '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'preference': -10},
268 '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
269 '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
270 '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
271 '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
272 '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
273 '138': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
274 '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
275 '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
278 '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 48, 'preference': -50},
279 '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 128, 'preference': -50},
280 '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'vcodec': 'none', 'abr': 256, 'preference': -50},
283 '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
284 '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
285 '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
286 '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
287 '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
288 '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'acodec': 'none', 'container': 'webm', 'vcodec': 'VP8', 'preference': -40},
289 '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
290 '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
291 '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
292 '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
293 '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
294 '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
295 '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
296 '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
297 '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'acodec': 'none', 'preference': -40},
300 '171': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 128, 'preference': -50},
301 '172': {'ext': 'webm', 'vcodec': 'none', 'format_note': 'DASH audio', 'abr': 256, 'preference': -50},
304 '_rtmp
': {'protocol': 'rtmp'},
310 'url
': 'http
://www
.youtube
.com
/watch?v
=BaW_jenozKc
',
314 'title
': 'youtube
-dl test video
"\'/\\ä↭𝕐',
315 'uploader': 'Philipp Hagemeister',
316 'uploader_id': 'phihag',
317 'upload_date': '20121002',
318 'description': 'test chars: "\'/\\ä↭𝕐
\ntest URL
: https
://github
.com
/rg3
/youtube
-dl
/issues
/1892\n\nThis
is a test video
for youtube
-dl
.\n\nFor more information
, contact phihag
@phihag.de .',
319 'categories
': ['Science
& Technology
'],
321 'dislike_count
': int,
325 'url
': 'http
://www
.youtube
.com
/watch?v
=UxxajLWwzqY
',
326 'note
': 'Test generic use_cipher_signature
video (#897)',
330 'upload_date': '20120506',
331 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
332 'description': 'md5:fea86fda2d5a5784273df5c7cc994d9f',
333 'uploader': 'Icona Pop',
334 'uploader_id': 'IconaPop',
338 'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
339 'note': 'Test VEVO video with age protection (#956)',
343 'upload_date': '20130703',
344 'title': 'Justin Timberlake - Tunnel Vision (Explicit)',
345 'description': 'md5:64249768eec3bc4276236606ea996373',
346 'uploader': 'justintimberlakeVEVO',
347 'uploader_id': 'justintimberlakeVEVO',
351 'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
352 'note': 'Embed-only video (#1746)',
356 'upload_date': '20120608',
357 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
358 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
359 'uploader': 'SET India',
360 'uploader_id': 'setindia'
364 'url': 'http://www.youtube.com/watch?v=a9LDPn-MO4I',
365 'note': '256k DASH audio (format 141) via DASH manifest',
369 'upload_date': '20121002',
370 'uploader_id': '8KVIDEO',
372 'uploader': '8KVIDEO',
373 'title': 'UHDTV TEST 8K VIDEO.mp4'
376 'youtube_include_dash_manifest': True,
380 # DASH manifest with encrypted signature
382 'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
386 'title': 'Afrojack - The Spark ft. Spree Wilson',
387 'description': 'md5:9717375db5a9a3992be4668bbf3bc0a8',
388 'uploader': 'AfrojackVEVO',
389 'uploader_id': 'AfrojackVEVO',
390 'upload_date': '20131011',
393 'youtube_include_dash_manifest': True,
399 def __init__(self
, *args
, **kwargs
):
400 super(YoutubeIE
, self
).__init
__(*args
, **kwargs
)
401 self
._player
_cache
= {}
403 def report_video_info_webpage_download(self
, video_id
):
404 """Report attempt to download video info webpage."""
405 self
.to_screen('%s: Downloading video info webpage' % video_id
)
407 def report_information_extraction(self
, video_id
):
408 """Report attempt to extract video information."""
409 self
.to_screen('%s: Extracting video information' % video_id
)
411 def report_unavailable_format(self
, video_id
, format
):
412 """Report extracted video URL."""
413 self
.to_screen('%s: Format %s not available' % (video_id
, format
))
415 def report_rtmp_download(self
):
416 """Indicate the download will use the RTMP protocol."""
417 self
.to_screen('RTMP download detected')
419 def _signature_cache_id(self
, example_sig
):
420 """ Return a string representation of a signature """
421 return '.'.join(compat_str(len(part
)) for part
in example_sig
.split('.'))
423 def _extract_signature_function(self
, video_id
, player_url
, example_sig
):
425 r
'.*-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.(?P<ext>[a-z]+)$',
428 raise ExtractorError('Cannot identify player %r' % player_url
)
429 player_type
= id_m
.group('ext')
430 player_id
= id_m
.group('id')
432 # Read from filesystem cache
433 func_id
= '%s_%s_%s' % (
434 player_type
, player_id
, self
._signature
_cache
_id
(example_sig
))
435 assert os
.path
.basename(func_id
) == func_id
437 cache_spec
= self
._downloader
.cache
.load('youtube-sigfuncs', func_id
)
438 if cache_spec
is not None:
439 return lambda s
: ''.join(s
[i
] for i
in cache_spec
)
441 if player_type
== 'js':
442 code
= self
._download
_webpage
(
443 player_url
, video_id
,
444 note
='Downloading %s player %s' % (player_type
, player_id
),
445 errnote
='Download of %s failed' % player_url
)
446 res
= self
._parse
_sig
_js
(code
)
447 elif player_type
== 'swf':
448 urlh
= self
._request
_webpage
(
449 player_url
, video_id
,
450 note
='Downloading %s player %s' % (player_type
, player_id
),
451 errnote
='Download of %s failed' % player_url
)
453 res
= self
._parse
_sig
_swf
(code
)
455 assert False, 'Invalid player type %r' % player_type
457 if cache_spec
is None:
458 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
459 cache_res
= res(test_string
)
460 cache_spec
= [ord(c
) for c
in cache_res
]
462 self
._downloader
.cache
.store('youtube-sigfuncs', func_id
, cache_spec
)
465 def _print_sig_code(self
, func
, example_sig
):
466 def gen_sig_code(idxs
):
467 def _genslice(start
, end
, step
):
468 starts
= '' if start
== 0 else str(start
)
469 ends
= (':%d' % (end
+step
)) if end
+ step
>= 0 else ':'
470 steps
= '' if step
== 1 else (':%d' % step
)
471 return 's[%s%s%s]' % (starts
, ends
, steps
)
474 start
= '(Never used)' # Quelch pyflakes warnings - start will be
475 # set as soon as step is set
476 for i
, prev
in zip(idxs
[1:], idxs
[:-1]):
480 yield _genslice(start
, prev
, step
)
483 if i
- prev
in [-1, 1]:
492 yield _genslice(start
, i
, step
)
494 test_string
= ''.join(map(compat_chr
, range(len(example_sig
))))
495 cache_res
= func(test_string
)
496 cache_spec
= [ord(c
) for c
in cache_res
]
497 expr_code
= ' + '.join(gen_sig_code(cache_spec
))
498 signature_id_tuple
= '(%s)' % (
499 ', '.join(compat_str(len(p
)) for p
in example_sig
.split('.')))
500 code
= ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
501 ' return %s\n') % (signature_id_tuple
, expr_code
)
502 self
.to_screen('Extracted signature function:\n' + code
)
504 def _parse_sig_js(self
, jscode
):
505 funcname
= self
._search
_regex
(
506 r
'signature=([$a-zA-Z]+)', jscode
,
507 'Initial JS player signature function name')
509 jsi
= JSInterpreter(jscode
)
510 initial_function
= jsi
.extract_function(funcname
)
511 return lambda s
: initial_function([s
])
513 def _parse_sig_swf(self
, file_contents
):
514 swfi
= SWFInterpreter(file_contents
)
515 TARGET_CLASSNAME
= 'SignatureDecipher'
516 searched_class
= swfi
.extract_class(TARGET_CLASSNAME
)
517 initial_function
= swfi
.extract_function(searched_class
, 'decipher')
518 return lambda s
: initial_function([s
])
520 def _decrypt_signature(self
, s
, video_id
, player_url
, age_gate
=False):
521 """Turn the encrypted s field into a working signature"""
523 if player_url
is None:
524 raise ExtractorError('Cannot decrypt signature without player_url')
526 if player_url
.startswith('//'):
527 player_url
= 'https:' + player_url
529 player_id
= (player_url
, self
._signature
_cache
_id
(s
))
530 if player_id
not in self
._player
_cache
:
531 func
= self
._extract
_signature
_function
(
532 video_id
, player_url
, s
534 self
._player
_cache
[player_id
] = func
535 func
= self
._player
_cache
[player_id
]
536 if self
._downloader
.params
.get('youtube_print_sig_code'):
537 self
._print
_sig
_code
(func
, s
)
539 except Exception as e
:
540 tb
= traceback
.format_exc()
541 raise ExtractorError(
542 'Signature extraction failed: ' + tb
, cause
=e
)
544 def _get_available_subtitles(self
, video_id
, webpage
):
546 sub_list
= self
._download
_webpage
(
547 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id
,
548 video_id
, note
=False)
549 except ExtractorError
as err
:
550 self
._downloader
.report_warning('unable to download video subtitles: %s' % compat_str(err
))
552 lang_list
= re
.findall(r
'name="([^"]*)"[^>]+lang_code="([\w\-]+)"', sub_list
)
557 if lang
in sub_lang_list
:
559 params
= compat_urllib_parse
.urlencode({
562 'fmt': self
._downloader
.params
.get('subtitlesformat', 'srt'),
563 'name': unescapeHTML(l
[0]).encode('utf-8'),
565 url
= 'https://www.youtube.com/api/timedtext?' + params
566 sub_lang_list
[lang
] = url
567 if not sub_lang_list
:
568 self
._downloader
.report_warning('video doesn\'t have subtitles')
572 def _get_available_automatic_caption(self
, video_id
, webpage
):
573 """We need the webpage for getting the captions url, pass it as an
574 argument to speed up the process."""
575 sub_format
= self
._downloader
.params
.get('subtitlesformat', 'srt')
576 self
.to_screen('%s: Looking for automatic captions' % video_id
)
577 mobj
= re
.search(r
';ytplayer.config = ({.*?});', webpage
)
578 err_msg
= 'Couldn\'t find automatic captions for %s' % video_id
580 self
._downloader
.report_warning(err_msg
)
582 player_config
= json
.loads(mobj
.group(1))
584 args
= player_config
[u
'args']
585 caption_url
= args
[u
'ttsurl']
586 timestamp
= args
[u
'timestamp']
587 # We get the available subtitles
588 list_params
= compat_urllib_parse
.urlencode({
593 list_url
= caption_url
+ '&' + list_params
594 caption_list
= self
._download
_xml
(list_url
, video_id
)
595 original_lang_node
= caption_list
.find('track')
596 if original_lang_node
is None or original_lang_node
.attrib
.get('kind') != 'asr' :
597 self
._downloader
.report_warning('Video doesn\'t have automatic captions')
599 original_lang
= original_lang_node
.attrib
['lang_code']
602 for lang_node
in caption_list
.findall('target'):
603 sub_lang
= lang_node
.attrib
['lang_code']
604 params
= compat_urllib_parse
.urlencode({
605 'lang': original_lang
,
611 sub_lang_list
[sub_lang
] = caption_url
+ '&' + params
613 # An extractor error can be raise by the download process if there are
614 # no automatic captions but there are subtitles
615 except (KeyError, ExtractorError
):
616 self
._downloader
.report_warning(err_msg
)
620 def extract_id(cls
, url
):
621 mobj
= re
.match(cls
._VALID
_URL
, url
, re
.VERBOSE
)
623 raise ExtractorError('Invalid URL: %s' % url
)
624 video_id
= mobj
.group(2)
627 def _extract_from_m3u8(self
, manifest_url
, video_id
):
629 def _get_urls(_manifest
):
630 lines
= _manifest
.split('\n')
631 urls
= filter(lambda l
: l
and not l
.startswith('#'),
634 manifest
= self
._download
_webpage
(manifest_url
, video_id
, 'Downloading formats manifest')
635 formats_urls
= _get_urls(manifest
)
636 for format_url
in formats_urls
:
637 itag
= self
._search
_regex
(r
'itag/(\d+?)/', format_url
, 'itag')
638 url_map
[itag
] = format_url
641 def _extract_annotations(self
, video_id
):
642 url
= 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
643 return self
._download
_webpage
(url
, video_id
, note
='Searching for annotations.', errnote
='Unable to download video annotations.')
645 def _real_extract(self
, url
):
647 'http' if self
._downloader
.params
.get('prefer_insecure', False)
650 # Extract original video URL from URL with redirection, like age verification, using next_url parameter
651 mobj
= re
.search(self
._NEXT
_URL
_RE
, url
)
653 url
= proto
+ '://www.youtube.com/' + compat_urllib_parse
.unquote(mobj
.group(1)).lstrip('/')
654 video_id
= self
.extract_id(url
)
657 url
= proto
+ '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1' % video_id
659 c
for c
in self
._downloader
.cookiejar
660 if c
.domain
== '.youtube.com' and c
.name
== 'PREF']
661 for pc
in pref_cookies
:
662 if 'hl=' in pc
.value
:
663 pc
.value
= re
.sub(r
'hl=[^&]+', 'hl=en', pc
.value
)
668 video_webpage
= self
._download
_webpage
(url
, video_id
)
670 # Attempt to extract SWF player URL
671 mobj
= re
.search(r
'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage
)
673 player_url
= re
.sub(r
'\\(.)', r
'\1', mobj
.group(1))
678 self
.report_video_info_webpage_download(video_id
)
679 if re
.search(r
'player-age-gate-content">', video_webpage
) is not None:
680 self
.report_age_confirmation()
682 # We simulate the access to the video from www.youtube.com/v/{video_id}
683 # this can be viewed without login into Youtube
684 data
= compat_urllib_parse
.urlencode({
685 'video_id': video_id
,
686 'eurl': 'https://youtube.googleapis.com/v/' + video_id
,
687 'sts': self
._search
_regex
(
688 r
'"sts"\s*:\s*(\d+)', video_webpage
, 'sts'),
690 video_info_url
= proto
+ '://www.youtube.com/get_video_info?' + data
691 video_info_webpage
= self
._download
_webpage
(video_info_url
, video_id
,
693 errnote
='unable to download video info webpage')
694 video_info
= compat_parse_qs(video_info_webpage
)
697 for el_type
in ['&el=embedded', '&el=detailpage', '&el=vevo', '']:
698 video_info_url
= (proto
+ '://www.youtube.com/get_video_info?&video_id=%s%s&ps=default&eurl=&gl=US&hl=en'
699 % (video_id
, el_type
))
700 video_info_webpage
= self
._download
_webpage
(video_info_url
, video_id
,
702 errnote
='unable to download video info webpage')
703 video_info
= compat_parse_qs(video_info_webpage
)
704 if 'token' in video_info
:
706 if 'token' not in video_info
:
707 if 'reason' in video_info
:
708 raise ExtractorError(
709 'YouTube said: %s' % video_info
['reason'][0],
710 expected
=True, video_id
=video_id
)
712 raise ExtractorError(
713 '"token" parameter not in video info for unknown reason',
716 if 'view_count' in video_info
:
717 view_count
= int(video_info
['view_count'][0])
721 # Check for "rental" videos
722 if 'ypc_video_rental_bar_text' in video_info
and 'author' not in video_info
:
723 raise ExtractorError('"rental" videos not supported')
725 # Start extracting information
726 self
.report_information_extraction(video_id
)
729 if 'author' not in video_info
:
730 raise ExtractorError('Unable to extract uploader name')
731 video_uploader
= compat_urllib_parse
.unquote_plus(video_info
['author'][0])
734 video_uploader_id
= None
735 mobj
= re
.search(r
'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage
)
737 video_uploader_id
= mobj
.group(1)
739 self
._downloader
.report_warning('unable to extract uploader nickname')
742 if 'title' in video_info
:
743 video_title
= video_info
['title'][0]
745 self
._downloader
.report_warning('Unable to extract video title')
749 # We try first to get a high quality image:
750 m_thumb
= re
.search(r
'<span itemprop="thumbnail".*?href="(.*?)">',
751 video_webpage
, re
.DOTALL
)
752 if m_thumb
is not None:
753 video_thumbnail
= m_thumb
.group(1)
754 elif 'thumbnail_url' not in video_info
:
755 self
._downloader
.report_warning('unable to extract video thumbnail')
756 video_thumbnail
= None
757 else: # don't panic if we can't find it
758 video_thumbnail
= compat_urllib_parse
.unquote_plus(video_info
['thumbnail_url'][0])
762 mobj
= re
.search(r
'(?s)id="eow-date.*?>(.*?)</span>', video_webpage
)
765 r
'(?s)id="watch-uploader-info".*?>.*?(?:Published|Uploaded|Streamed live) on (.*?)</strong>',
768 upload_date
= ' '.join(re
.sub(r
'[/,-]', r
' ', mobj
.group(1)).split())
769 upload_date
= unified_strdate(upload_date
)
771 m_cat_container
= self
._search
_regex
(
772 r
'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
773 video_webpage
, 'categories', fatal
=False)
775 category
= self
._html
_search
_regex
(
776 r
'(?s)<a[^<]+>(.*?)</a>', m_cat_container
, 'category',
778 video_categories
= None if category
is None else [category
]
780 video_categories
= None
783 video_description
= get_element_by_id("eow-description", video_webpage
)
784 if video_description
:
785 video_description
= re
.sub(r
'''(?x)
787 (?:[a-zA-Z-]+="[^"]+"\s+)*?
789 (?:[a-zA-Z-]+="[^"]+"\s+)*?
790 class="yt-uix-redirect-link"\s*>
793 ''', r
'\1', video_description
)
794 video_description
= clean_html(video_description
)
796 fd_mobj
= re
.search(r
'<meta name="description" content="([^"]+)"', video_webpage
)
798 video_description
= unescapeHTML(fd_mobj
.group(1))
800 video_description
= ''
802 def _extract_count(count_name
):
803 count
= self
._search
_regex
(
804 r
'id="watch-%s"[^>]*>.*?([\d,]+)\s*</span>' % re
.escape(count_name
),
805 video_webpage
, count_name
, default
=None)
806 if count
is not None:
807 return int(count
.replace(',', ''))
809 like_count
= _extract_count('like')
810 dislike_count
= _extract_count('dislike')
813 video_subtitles
= self
.extract_subtitles(video_id
, video_webpage
)
815 if self
._downloader
.params
.get('listsubtitles', False):
816 self
._list
_available
_subtitles
(video_id
, video_webpage
)
819 if 'length_seconds' not in video_info
:
820 self
._downloader
.report_warning('unable to extract video duration')
821 video_duration
= None
823 video_duration
= int(compat_urllib_parse
.unquote_plus(video_info
['length_seconds'][0]))
826 video_annotations
= None
827 if self
._downloader
.params
.get('writeannotations', False):
828 video_annotations
= self
._extract
_annotations
(video_id
)
830 # Decide which formats to download
832 mobj
= re
.search(r
';ytplayer\.config\s*=\s*({.*?});', video_webpage
)
834 raise ValueError('Could not find vevo ID')
835 json_code
= uppercase_escape(mobj
.group(1))
836 ytplayer_config
= json
.loads(json_code
)
837 args
= ytplayer_config
['args']
838 # Easy way to know if the 's' value is in url_encoded_fmt_stream_map
839 # this signatures are encrypted
840 if 'url_encoded_fmt_stream_map' not in args
:
841 raise ValueError('No stream_map present') # caught below
842 re_signature
= re
.compile(r
'[&,]s=')
843 m_s
= re_signature
.search(args
['url_encoded_fmt_stream_map'])
845 self
.to_screen('%s: Encrypted signatures detected.' % video_id
)
846 video_info
['url_encoded_fmt_stream_map'] = [args
['url_encoded_fmt_stream_map']]
847 m_s
= re_signature
.search(args
.get('adaptive_fmts', ''))
849 if 'adaptive_fmts' in video_info
:
850 video_info
['adaptive_fmts'][0] += ',' + args
['adaptive_fmts']
852 video_info
['adaptive_fmts'] = [args
['adaptive_fmts']]
856 def _map_to_format_list(urlmap
):
858 for itag
, video_real_url
in urlmap
.items():
861 'url': video_real_url
,
862 'player_url': player_url
,
864 if itag
in self
._formats
:
865 dct
.update(self
._formats
[itag
])
869 if 'conn' in video_info
and video_info
['conn'][0].startswith('rtmp'):
870 self
.report_rtmp_download()
872 'format_id': '_rtmp',
874 'url': video_info
['conn'][0],
875 'player_url': player_url
,
877 elif len(video_info
.get('url_encoded_fmt_stream_map', [])) >= 1 or len(video_info
.get('adaptive_fmts', [])) >= 1:
878 encoded_url_map
= video_info
.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info
.get('adaptive_fmts',[''])[0]
879 if 'rtmpe%3Dyes' in encoded_url_map
:
880 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected
=True)
882 for url_data_str
in encoded_url_map
.split(','):
883 url_data
= compat_parse_qs(url_data_str
)
884 if 'itag' not in url_data
or 'url' not in url_data
:
886 format_id
= url_data
['itag'][0]
887 url
= url_data
['url'][0]
889 if 'sig' in url_data
:
890 url
+= '&signature=' + url_data
['sig'][0]
891 elif 's' in url_data
:
892 encrypted_sig
= url_data
['s'][0]
895 jsplayer_url_json
= self
._search
_regex
(
896 r
'"assets":.+?"js":\s*("[^"]+")',
897 video_webpage
, 'JS player URL')
898 player_url
= json
.loads(jsplayer_url_json
)
899 if player_url
is None:
900 player_url_json
= self
._search
_regex
(
901 r
'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
902 video_webpage
, 'age gate player URL')
903 player_url
= json
.loads(player_url_json
)
905 if self
._downloader
.params
.get('verbose'):
906 if player_url
is None:
907 player_version
= 'unknown'
908 player_desc
= 'unknown'
910 if player_url
.endswith('swf'):
911 player_version
= self
._search
_regex
(
912 r
'-(.+?)(?:/watch_as3)?\.swf$', player_url
,
913 'flash player', fatal
=False)
914 player_desc
= 'flash player %s' % player_version
916 player_version
= self
._search
_regex
(
917 r
'html5player-([^/]+?)(?:/html5player)?\.js',
919 'html5 player', fatal
=False)
920 player_desc
= 'html5 player %s' % player_version
922 parts_sizes
= self
._signature
_cache
_id
(encrypted_sig
)
923 self
.to_screen('{%s} signature length %s, %s' %
924 (format_id
, parts_sizes
, player_desc
))
926 signature
= self
._decrypt
_signature
(
927 encrypted_sig
, video_id
, player_url
, age_gate
)
928 url
+= '&signature=' + signature
929 if 'ratebypass' not in url
:
930 url
+= '&ratebypass=yes'
931 url_map
[format_id
] = url
932 formats
= _map_to_format_list(url_map
)
933 elif video_info
.get('hlsvp'):
934 manifest_url
= video_info
['hlsvp'][0]
935 url_map
= self
._extract
_from
_m
3u8(manifest_url
, video_id
)
936 formats
= _map_to_format_list(url_map
)
938 raise ExtractorError('no conn, hlsvp or url_encoded_fmt_stream_map information found in video info')
940 # Look for the DASH manifest
941 if (self
._downloader
.params
.get('youtube_include_dash_manifest', False)):
943 # The DASH manifest used needs to be the one from the original video_webpage.
944 # The one found in get_video_info seems to be using different signatures.
945 # However, in the case of an age restriction there won't be any embedded dashmpd in the video_webpage.
946 # Luckily, it seems, this case uses some kind of default signature (len == 86), so the
947 # combination of get_video_info and the _static_decrypt_signature() decryption fallback will work here.
949 dash_manifest_url
= video_info
.get('dashmpd')[0]
951 dash_manifest_url
= ytplayer_config
['args']['dashmpd']
952 def decrypt_sig(mobj
):
954 dec_s
= self
._decrypt
_signature
(s
, video_id
, player_url
, age_gate
)
955 return '/signature/%s' % dec_s
956 dash_manifest_url
= re
.sub(r
'/s/([\w\.]+)', decrypt_sig
, dash_manifest_url
)
957 dash_doc
= self
._download
_xml
(
958 dash_manifest_url
, video_id
,
959 note
='Downloading DASH manifest',
960 errnote
='Could not download DASH manifest')
961 for r
in dash_doc
.findall('.//{urn:mpeg:DASH:schema:MPD:2011}Representation'):
962 url_el
= r
.find('{urn:mpeg:DASH:schema:MPD:2011}BaseURL')
965 format_id
= r
.attrib
['id']
966 video_url
= url_el
.text
967 filesize
= int_or_none(url_el
.attrib
.get('{http://youtube.com/yt/2012/10/10}contentLength'))
969 'format_id': format_id
,
971 'width': int_or_none(r
.attrib
.get('width')),
972 'tbr': int_or_none(r
.attrib
.get('bandwidth'), 1000),
973 'asr': int_or_none(r
.attrib
.get('audioSamplingRate')),
974 'filesize': filesize
,
977 existing_format
= next(
979 if fo
['format_id'] == format_id
)
980 except StopIteration:
981 f
.update(self
._formats
.get(format_id
, {}))
984 existing_format
.update(f
)
986 except (ExtractorError
, KeyError) as e
:
987 self
.report_warning('Skipping DASH manifest: %s' % e
, video_id
)
989 self
._sort
_formats
(formats
)
993 'uploader': video_uploader
,
994 'uploader_id': video_uploader_id
,
995 'upload_date': upload_date
,
996 'title': video_title
,
997 'thumbnail': video_thumbnail
,
998 'description': video_description
,
999 'categories': video_categories
,
1000 'subtitles': video_subtitles
,
1001 'duration': video_duration
,
1002 'age_limit': 18 if age_gate
else 0,
1003 'annotations': video_annotations
,
1004 'webpage_url': proto
+ '://www.youtube.com/watch?v=%s' % video_id
,
1005 'view_count': view_count
,
1006 'like_count': like_count
,
1007 'dislike_count': dislike_count
,
1011 class YoutubePlaylistIE(YoutubeBaseInfoExtractor
):
1012 IE_DESC
= 'YouTube.com playlists'
1013 _VALID_URL
= r
"""(?x)(?:
1018 (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/videoseries)
1019 \? (?:.*?&)*? (?:p|a|list)=
1023 (?:PL|LL|EC|UU|FL|RD)?[0-9A-Za-z-_]{10,}
1024 # Top tracks, they can also include dots
1029 ((?:PL|LL|EC|UU|FL|RD)[0-9A-Za-z-_]{10,})
1031 _TEMPLATE_URL
= 'https://www.youtube.com/playlist?list=%s'
1032 _MORE_PAGES_INDICATOR
= r
'data-link-type="next"'
1033 _VIDEO_RE
= r
'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&[^"]*?index=(?P<index>\d+)'
1034 IE_NAME
= 'youtube:playlist'
1036 'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
1038 'title': 'ytdl test PL',
1040 'playlist_count': 3,
1042 'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
1044 'title': 'YDL_Empty_List',
1046 'playlist_count': 0,
1048 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
1049 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
1051 'title': '29C3: Not my department',
1053 'playlist_count': 95,
1055 'note': 'issue #673',
1056 'url': 'PLBB231211A4F62143',
1058 'title': 'Team Fortress 2 (Class-based LP)',
1060 'playlist_mincount': 26,
1062 'note': 'Large playlist',
1063 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
1065 'title': 'Uploads from Cauchemar',
1067 'playlist_mincount': 799,
1069 'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
1071 'title': 'YDL_safe_search',
1073 'playlist_count': 2,
1076 'url': 'http://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
1077 'playlist_count': 4,
1082 'note': 'Embedded SWF player',
1083 'url': 'http://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
1084 'playlist_count': 4,
1090 def _real_initialize(self
):
1093 def _ids_to_results(self
, ids
):
1095 self
.url_result(vid_id
, 'Youtube', video_id
=vid_id
)
1098 def _extract_mix(self
, playlist_id
):
1099 # The mixes are generated from a a single video
1100 # the id of the playlist is just 'RD' + video_id
1101 url
= 'https://youtube.com/watch?v=%s&list=%s' % (playlist_id
[-11:], playlist_id
)
1102 webpage
= self
._download
_webpage
(
1103 url
, playlist_id
, 'Downloading Youtube mix')
1104 search_title
= lambda class_name
: get_element_by_attribute('class', class_name
, webpage
)
1106 search_title('playlist-title') or
1107 search_title('title long-title') or
1108 search_title('title'))
1109 title
= clean_html(title_span
)
1110 ids
= orderedSet(re
.findall(
1111 r
'''(?xs)data-video-username=".*?".*?
1112 href="/watch\?v=([0-9A-Za-z_-]{11})&[^"]*?list=%s''' % re
.escape(playlist_id
),
1114 url_results
= self
._ids
_to
_results
(ids
)
1116 return self
.playlist_result(url_results
, playlist_id
, title
)
1118 def _real_extract(self
, url
):
1119 # Extract playlist id
1120 mobj
= re
.match(self
._VALID
_URL
, url
)
1122 raise ExtractorError('Invalid URL: %s' % url
)
1123 playlist_id
= mobj
.group(1) or mobj
.group(2)
1125 # Check if it's a video-specific URL
1126 query_dict
= compat_urlparse
.parse_qs(compat_urlparse
.urlparse(url
).query
)
1127 if 'v' in query_dict
:
1128 video_id
= query_dict
['v'][0]
1129 if self
._downloader
.params
.get('noplaylist'):
1130 self
.to_screen('Downloading just video %s because of --no-playlist' % video_id
)
1131 return self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1133 self
.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id
, video_id
))
1135 if playlist_id
.startswith('RD'):
1136 # Mixes require a custom extraction process
1137 return self
._extract
_mix
(playlist_id
)
1138 if playlist_id
.startswith('TL'):
1139 raise ExtractorError('For downloading YouTube.com top lists, use '
1140 'the "yttoplist" keyword, for example "youtube-dl \'yttoplist:music:Top Tracks\'"', expected
=True)
1142 url
= self
._TEMPLATE
_URL
% playlist_id
1143 page
= self
._download
_webpage
(url
, playlist_id
)
1144 more_widget_html
= content_html
= page
1146 # Check if the playlist exists or is private
1147 if re
.search(r
'<div class="yt-alert-message">[^<]*?(The|This) playlist (does not exist|is private)[^<]*?</div>', page
) is not None:
1148 raise ExtractorError(
1149 'The playlist doesn\'t exist or is private, use --username or '
1150 '--netrc to access it.',
1153 # Extract the video ids from the playlist pages
1156 for page_num
in itertools
.count(1):
1157 matches
= re
.finditer(self
._VIDEO
_RE
, content_html
)
1158 # We remove the duplicates and the link with index 0
1159 # (it's not the first video of the playlist)
1160 new_ids
= orderedSet(m
.group('id') for m
in matches
if m
.group('index') != '0')
1163 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1167 more
= self
._download
_json
(
1168 'https://youtube.com/%s' % mobj
.group('more'), playlist_id
,
1169 'Downloading page #%s' % page_num
,
1170 transform_source
=uppercase_escape
)
1171 content_html
= more
['content_html']
1172 more_widget_html
= more
['load_more_widget_html']
1174 playlist_title
= self
._html
_search
_regex
(
1175 r
'(?s)<h1 class="pl-header-title[^"]*">\s*(.*?)\s*</h1>',
1178 url_results
= self
._ids
_to
_results
(ids
)
1179 return self
.playlist_result(url_results
, playlist_id
, playlist_title
)
1182 class YoutubeTopListIE(YoutubePlaylistIE
):
1183 IE_NAME
= 'youtube:toplist'
1184 IE_DESC
= ('YouTube.com top lists, "yttoplist:{channel}:{list title}"'
1185 ' (Example: "yttoplist:music:Top Tracks")')
1186 _VALID_URL
= r
'yttoplist:(?P<chann>.*?):(?P<title>.*?)$'
1188 'url': 'yttoplist:music:Trending',
1189 'playlist_mincount': 5,
1190 'skip': 'Only works for logged-in users',
1193 def _real_extract(self
, url
):
1194 mobj
= re
.match(self
._VALID
_URL
, url
)
1195 channel
= mobj
.group('chann')
1196 title
= mobj
.group('title')
1197 query
= compat_urllib_parse
.urlencode({'title': title}
)
1198 channel_page
= self
._download
_webpage
(
1199 'https://www.youtube.com/%s' % channel
, title
)
1200 link
= self
._html
_search
_regex
(
1202 <a\s+href="([^"]+)".*?>\s*
1203 <span\s+class="branded-page-module-title-text">\s*
1204 <span[^>]*>.*?%s.*?</span>''' % re
.escape(query
),
1205 channel_page
, 'list')
1206 url
= compat_urlparse
.urljoin('https://www.youtube.com/', link
)
1208 video_re
= r
'data-index="\d+".*?data-video-id="([0-9A-Za-z_-]{11})"'
1210 # sometimes the webpage doesn't contain the videos
1211 # retry until we get them
1212 for i
in itertools
.count(0):
1213 msg
= 'Downloading Youtube mix'
1215 msg
+= ', retry #%d' % i
1217 webpage
= self
._download
_webpage
(url
, title
, msg
)
1218 ids
= orderedSet(re
.findall(video_re
, webpage
))
1221 url_results
= self
._ids
_to
_results
(ids
)
1222 return self
.playlist_result(url_results
, playlist_title
=title
)
1225 class YoutubeChannelIE(InfoExtractor
):
1226 IE_DESC
= 'YouTube.com channels'
1227 _VALID_URL
= r
"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)"
1228 _MORE_PAGES_INDICATOR
= 'yt-uix-load-more'
1229 _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'
1230 IE_NAME
= 'youtube:channel'
1232 'note': 'paginated channel',
1233 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
1234 'playlist_mincount': 91,
1237 def extract_videos_from_page(self
, page
):
1239 for mobj
in re
.finditer(r
'href="/watch\?v=([0-9A-Za-z_-]+)&?', page
):
1240 if mobj
.group(1) not in ids_in_page
:
1241 ids_in_page
.append(mobj
.group(1))
1244 def _real_extract(self
, url
):
1245 # Extract channel id
1246 mobj
= re
.match(self
._VALID
_URL
, url
)
1248 raise ExtractorError('Invalid URL: %s' % url
)
1250 # Download channel page
1251 channel_id
= mobj
.group(1)
1253 url
= 'https://www.youtube.com/channel/%s/videos' % channel_id
1254 channel_page
= self
._download
_webpage
(url
, channel_id
)
1255 autogenerated
= re
.search(r
'''(?x)
1257 channel-header-autogenerated-label|
1258 yt-channel-title-autogenerated
1259 )[^"]*"''', channel_page
) is not None
1262 # The videos are contained in a single page
1263 # the ajax pages can't be used, they are empty
1264 video_ids
= self
.extract_videos_from_page(channel_page
)
1266 # Download all channel pages using the json-based channel_ajax query
1267 for pagenum
in itertools
.count(1):
1268 url
= self
._MORE
_PAGES
_URL
% (pagenum
, channel_id
)
1269 page
= self
._download
_json
(
1270 url
, channel_id
, note
='Downloading page #%s' % pagenum
,
1271 transform_source
=uppercase_escape
)
1273 ids_in_page
= self
.extract_videos_from_page(page
['content_html'])
1274 video_ids
.extend(ids_in_page
)
1276 if self
._MORE
_PAGES
_INDICATOR
not in page
['load_more_widget_html']:
1279 self
._downloader
.to_screen('[youtube] Channel %s: Found %i videos' % (channel_id
, len(video_ids
)))
1281 url_entries
= [self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1282 for video_id
in video_ids
]
1283 return self
.playlist_result(url_entries
, channel_id
)
1286 class YoutubeUserIE(InfoExtractor
):
1287 IE_DESC
= 'YouTube.com user videos (URL or "ytuser" keyword)'
1288 _VALID_URL
= r
'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)([A-Za-z0-9_-]+)'
1289 _TEMPLATE_URL
= 'https://gdata.youtube.com/feeds/api/users/%s'
1290 _GDATA_PAGE_SIZE
= 50
1291 _GDATA_URL
= 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
1292 IE_NAME
= 'youtube:user'
1295 'url': 'https://www.youtube.com/user/TheLinuxFoundation',
1296 'playlist_mincount': 320,
1298 'title': 'TheLinuxFoundation',
1301 'url': 'ytuser:phihag',
1302 'only_matching': True,
1306 def suitable(cls
, url
):
1307 # Don't return True if the url can be extracted with other youtube
1308 # extractor, the regex would is too permissive and it would match.
1309 other_ies
= iter(klass
for (name
, klass
) in globals().items() if name
.endswith('IE') and klass
is not cls
)
1310 if any(ie
.suitable(url
) for ie
in other_ies
): return False
1311 else: return super(YoutubeUserIE
, cls
).suitable(url
)
1313 def _real_extract(self
, url
):
1315 mobj
= re
.match(self
._VALID
_URL
, url
)
1317 raise ExtractorError('Invalid URL: %s' % url
)
1319 username
= mobj
.group(1)
1321 # Download video ids using YouTube Data API. Result size per
1322 # query is limited (currently to 50 videos) so we need to query
1323 # page by page until there are no video ids - it means we got
1326 def download_page(pagenum
):
1327 start_index
= pagenum
* self
._GDATA
_PAGE
_SIZE
+ 1
1329 gdata_url
= self
._GDATA
_URL
% (username
, self
._GDATA
_PAGE
_SIZE
, start_index
)
1330 page
= self
._download
_webpage
(
1331 gdata_url
, username
,
1332 'Downloading video ids from %d to %d' % (
1333 start_index
, start_index
+ self
._GDATA
_PAGE
_SIZE
))
1336 response
= json
.loads(page
)
1337 except ValueError as err
:
1338 raise ExtractorError('Invalid JSON in API response: ' + compat_str(err
))
1339 if 'entry' not in response
['feed']:
1342 # Extract video identifiers
1343 entries
= response
['feed']['entry']
1344 for entry
in entries
:
1345 title
= entry
['title']['$t']
1346 video_id
= entry
['id']['$t'].split('/')[-1]
1350 'ie_key': 'Youtube',
1354 url_results
= OnDemandPagedList(download_page
, self
._GDATA
_PAGE
_SIZE
)
1356 return self
.playlist_result(url_results
, playlist_title
=username
)
1359 class YoutubeSearchIE(SearchInfoExtractor
):
1360 IE_DESC
= 'YouTube.com searches'
1361 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc'
1363 IE_NAME
= 'youtube:search'
1364 _SEARCH_KEY
= 'ytsearch'
1366 def _get_n_results(self
, query
, n
):
1367 """Get a specified number of results for a query"""
1374 while (PAGE_SIZE
* pagenum
) < limit
:
1375 result_url
= self
._API
_URL
% (
1376 compat_urllib_parse
.quote_plus(query
.encode('utf-8')),
1377 (PAGE_SIZE
* pagenum
) + 1)
1378 data_json
= self
._download
_webpage
(
1379 result_url
, video_id
='query "%s"' % query
,
1380 note
='Downloading page %s' % (pagenum
+ 1),
1381 errnote
='Unable to download API page')
1382 data
= json
.loads(data_json
)
1383 api_response
= data
['data']
1385 if 'items' not in api_response
:
1386 raise ExtractorError(
1387 '[youtube] No video results', expected
=True)
1389 new_ids
= list(video
['id'] for video
in api_response
['items'])
1390 video_ids
+= new_ids
1392 limit
= min(n
, api_response
['totalItems'])
1395 if len(video_ids
) > n
:
1396 video_ids
= video_ids
[:n
]
1397 videos
= [self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1398 for video_id
in video_ids
]
1399 return self
.playlist_result(videos
, query
)
1402 class YoutubeSearchDateIE(YoutubeSearchIE
):
1403 IE_NAME
= YoutubeSearchIE
.IE_NAME
+ ':date'
1404 _API_URL
= 'https://gdata.youtube.com/feeds/api/videos?q=%s&start-index=%i&max-results=50&v=2&alt=jsonc&orderby=published'
1405 _SEARCH_KEY
= 'ytsearchdate'
1406 IE_DESC
= 'YouTube.com searches, newest videos first'
1409 class YoutubeSearchURLIE(InfoExtractor
):
1410 IE_DESC
= 'YouTube.com search URLs'
1411 IE_NAME
= 'youtube:search_url'
1412 _VALID_URL
= r
'https?://(?:www\.)?youtube\.com/results\?(.*?&)?search_query=(?P<query>[^&]+)(?:[&]|$)'
1414 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
1415 'playlist_mincount': 5,
1417 'title': 'youtube-dl test video',
1421 def _real_extract(self
, url
):
1422 mobj
= re
.match(self
._VALID
_URL
, url
)
1423 query
= compat_urllib_parse
.unquote_plus(mobj
.group('query'))
1425 webpage
= self
._download
_webpage
(url
, query
)
1426 result_code
= self
._search
_regex
(
1427 r
'(?s)<ol class="item-section"(.*?)</ol>', webpage
, 'result HTML')
1429 part_codes
= re
.findall(
1430 r
'(?s)<h3 class="yt-lockup-title">(.*?)</h3>', result_code
)
1432 for part_code
in part_codes
:
1433 part_title
= self
._html
_search
_regex
(
1434 [r
'(?s)title="([^"]+)"', r
'>([^<]+)</a>'], part_code
, 'item title', fatal
=False)
1435 part_url_snippet
= self
._html
_search
_regex
(
1436 r
'(?s)href="([^"]+)"', part_code
, 'item URL')
1437 part_url
= compat_urlparse
.urljoin(
1438 'https://www.youtube.com/', part_url_snippet
)
1442 'title': part_title
,
1446 '_type': 'playlist',
1452 class YoutubeShowIE(InfoExtractor
):
1453 IE_DESC
= 'YouTube.com (multi-season) shows'
1454 _VALID_URL
= r
'https?://www\.youtube\.com/show/(?P<id>[^?#]*)'
1455 IE_NAME
= 'youtube:show'
1457 'url': 'http://www.youtube.com/show/airdisasters',
1458 'playlist_mincount': 3,
1460 'id': 'airdisasters',
1461 'title': 'Air Disasters',
1465 def _real_extract(self
, url
):
1466 mobj
= re
.match(self
._VALID
_URL
, url
)
1467 playlist_id
= mobj
.group('id')
1468 webpage
= self
._download
_webpage
(
1469 url
, playlist_id
, 'Downloading show webpage')
1470 # There's one playlist for each season of the show
1471 m_seasons
= list(re
.finditer(r
'href="(/playlist\?list=.*?)"', webpage
))
1472 self
.to_screen('%s: Found %s seasons' % (playlist_id
, len(m_seasons
)))
1475 'https://www.youtube.com' + season
.group(1), 'YoutubePlaylist')
1476 for season
in m_seasons
1478 title
= self
._og
_search
_title
(webpage
, fatal
=False)
1481 '_type': 'playlist',
1488 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor
):
1490 Base class for extractors that fetch info from
1491 http://www.youtube.com/feed_ajax
1492 Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
1494 _LOGIN_REQUIRED
= True
1495 # use action_load_personal_feed instead of action_load_system_feed
1496 _PERSONAL_FEED
= False
1499 def _FEED_TEMPLATE(self
):
1500 action
= 'action_load_system_feed'
1501 if self
._PERSONAL
_FEED
:
1502 action
= 'action_load_personal_feed'
1503 return 'https://www.youtube.com/feed_ajax?%s=1&feed_name=%s&paging=%%s' % (action
, self
._FEED
_NAME
)
1507 return 'youtube:%s' % self
._FEED
_NAME
1509 def _real_initialize(self
):
1512 def _real_extract(self
, url
):
1515 for i
in itertools
.count(1):
1516 info
= self
._download
_json
(self
._FEED
_TEMPLATE
% paging
,
1517 '%s feed' % self
._FEED
_NAME
,
1518 'Downloading page %s' % i
)
1519 feed_html
= info
.get('feed_html') or info
.get('content_html')
1520 load_more_widget_html
= info
.get('load_more_widget_html') or feed_html
1521 m_ids
= re
.finditer(r
'"/watch\?v=(.*?)["&]', feed_html
)
1522 ids
= orderedSet(m
.group(1) for m
in m_ids
)
1523 feed_entries
.extend(
1524 self
.url_result(video_id
, 'Youtube', video_id
=video_id
)
1525 for video_id
in ids
)
1527 r
'data-uix-load-more-href="/?[^"]+paging=(?P<paging>\d+)',
1528 load_more_widget_html
)
1531 paging
= mobj
.group('paging')
1532 return self
.playlist_result(feed_entries
, playlist_title
=self
._PLAYLIST
_TITLE
)
1534 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor
):
1535 IE_DESC
= 'YouTube.com recommended videos, "ytrec" keyword (requires authentication)'
1536 _VALID_URL
= r
'https?://www\.youtube\.com/feed/recommended|:ytrec(?:ommended)?'
1537 _FEED_NAME
= 'recommended'
1538 _PLAYLIST_TITLE
= 'Youtube Recommended videos'
1540 class YoutubeWatchLaterIE(YoutubeFeedsInfoExtractor
):
1541 IE_DESC
= 'Youtube watch later list, "ytwatchlater" keyword (requires authentication)'
1542 _VALID_URL
= r
'https?://www\.youtube\.com/feed/watch_later|:ytwatchlater'
1543 _FEED_NAME
= 'watch_later'
1544 _PLAYLIST_TITLE
= 'Youtube Watch Later'
1545 _PERSONAL_FEED
= True
1547 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor
):
1548 IE_DESC
= 'Youtube watch history, "ythistory" keyword (requires authentication)'
1549 _VALID_URL
= 'https?://www\.youtube\.com/feed/history|:ythistory'
1550 _FEED_NAME
= 'history'
1551 _PERSONAL_FEED
= True
1552 _PLAYLIST_TITLE
= 'Youtube Watch History'
1554 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor
):
1555 IE_NAME
= 'youtube:favorites'
1556 IE_DESC
= 'YouTube.com favourite videos, "ytfav" keyword (requires authentication)'
1557 _VALID_URL
= r
'https?://www\.youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
1558 _LOGIN_REQUIRED
= True
1560 def _real_extract(self
, url
):
1561 webpage
= self
._download
_webpage
('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
1562 playlist_id
= self
._search
_regex
(r
'list=(.+?)["&]', webpage
, 'favourites playlist id')
1563 return self
.url_result(playlist_id
, 'YoutubePlaylist')
1566 class YoutubeSubscriptionsIE(YoutubePlaylistIE
):
1567 IE_NAME
= 'youtube:subscriptions'
1568 IE_DESC
= 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
1569 _VALID_URL
= r
'https?://www\.youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
1572 def _real_extract(self
, url
):
1573 title
= 'Youtube Subscriptions'
1574 page
= self
._download
_webpage
('https://www.youtube.com/feed/subscriptions', title
)
1576 # The extraction process is the same as for playlists, but the regex
1577 # for the video ids doesn't contain an index
1579 more_widget_html
= content_html
= page
1581 for page_num
in itertools
.count(1):
1582 matches
= re
.findall(r
'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html
)
1583 new_ids
= orderedSet(matches
)
1586 mobj
= re
.search(r
'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html
)
1590 more
= self
._download
_json
(
1591 'https://youtube.com/%s' % mobj
.group('more'), title
,
1592 'Downloading page #%s' % page_num
,
1593 transform_source
=uppercase_escape
)
1594 content_html
= more
['content_html']
1595 more_widget_html
= more
['load_more_widget_html']
1598 '_type': 'playlist',
1600 'entries': self
._ids
_to
_results
(ids
),
1604 class YoutubeTruncatedURLIE(InfoExtractor
):
1605 IE_NAME
= 'youtube:truncated_url'
1606 IE_DESC
= False # Do not list
1607 _VALID_URL
= r
'''(?x)
1608 (?:https?://)?[^/]+/watch\?(?:
1610 annotation_id=annotation_[^&]+
1612 (?:https?://)?(?:www\.)?youtube\.com/attribution_link\?a=[^&]+$
1616 'url': 'http://www.youtube.com/watch?annotation_id=annotation_3951667041',
1617 'only_matching': True,
1619 'url': 'http://www.youtube.com/watch?',
1620 'only_matching': True,
1623 def _real_extract(self
, url
):
1624 raise ExtractorError(
1625 'Did you forget to quote the URL? Remember that & is a meta '
1626 'character in most shells, so you want to put the URL in quotes, '
1628 '"http://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
1629 ' or simply youtube-dl BaW_jenozKc .',