]>
jfr.im git - yt-dlp.git/blob - youtube_dl/extractor/vk.py
2 from __future__
import unicode_literals
8 from .common
import InfoExtractor
9 from ..compat
import compat_urlparse
24 from .dailymotion
import DailymotionIE
25 from .pladform
import PladformIE
26 from .vimeo
import VimeoIE
27 from .youtube
import YoutubeIE
30 class VKBaseIE(InfoExtractor
):
34 username
, password
= self
._get
_login
_info
()
38 login_page
, url_handle
= self
._download
_webpage
_handle
(
39 'https://vk.com', None, 'Downloading login page')
41 login_form
= self
._hidden
_inputs
(login_page
)
44 'email': username
.encode('cp1251'),
45 'pass': password
.encode('cp1251'),
48 # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
49 # and expects the first one to be set rather than second (see
50 # https://github.com/ytdl-org/youtube-dl/issues/9841#issuecomment-227871201).
51 # As of RFC6265 the newer one cookie should be set into cookie store
52 # what actually happens.
53 # We will workaround this VK issue by resetting the remixlhk cookie to
54 # the first one manually.
55 for header
, cookies
in url_handle
.headers
.items():
56 if header
.lower() != 'set-cookie':
58 if sys
.version_info
[0] >= 3:
59 cookies
= cookies
.encode('iso-8859-1')
60 cookies
= cookies
.decode('utf-8')
61 remixlhk
= re
.search(r
'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies
)
63 value
, domain
= remixlhk
.groups()
64 self
._set
_cookie
(domain
, 'remixlhk', value
)
67 login_page
= self
._download
_webpage
(
68 'https://login.vk.com/?act=login', None,
70 data
=urlencode_postdata(login_form
))
72 if re
.search(r
'onLoginFailed', login_page
):
74 'Unable to login, incorrect username and/or password', expected
=True)
76 def _real_initialize(self
):
87 (?:(?:m|new)\.)?vk\.com/video_|
90 ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
92 (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
93 (?:www\.)?daxab.com/embed/
95 (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
100 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
101 'md5': '7babad3b85ea2e91948005b1b8b0cb84',
103 'id': '-77521_162222515',
105 'title': 'ProtivoGunz - Хуёвая песня',
106 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
107 'uploader_id': '-77521',
109 'timestamp': 1329049880,
110 'upload_date': '20120212',
114 'url': 'http://vk.com/video205387401_165548505',
115 'md5': '6c0aeb2e90396ba97035b9cbde548700',
117 'id': '205387401_165548505',
120 'uploader': 'Tom Cruise',
121 'uploader_id': '205387401',
123 'timestamp': 1374364108,
124 'upload_date': '20130720',
128 'note': 'Embedded video',
129 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
130 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
132 'id': '32194266_162925554',
134 'uploader': 'Vladimir Gavrin',
137 'upload_date': '20120730',
140 'skip': 'This video has been removed from public access.',
144 # please update if you find a video whose URL follows the same pattern
145 'url': 'http://vk.com/video-8871596_164049491',
146 'md5': 'a590bcaf3d543576c9bd162812387666',
147 'note': 'Only available for registered users',
149 'id': '-8871596_164049491',
151 'uploader': 'Триллеры',
152 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
154 'upload_date': '20121218',
157 'skip': 'Requires vk account credentials',
160 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
161 'md5': '4d7a5ef8cf114dfa09577e57b2993202',
163 'id': '-43215063_168067957',
165 'uploader': 'Киномания - лучшее из мира кино',
168 'upload_date': '20140328',
170 'skip': 'Requires vk account credentials',
173 'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
174 'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
175 'note': 'ivi.ru embed',
177 'id': '-43215063_169084319',
179 'title': 'Книга Илая',
181 'upload_date': '20140626',
184 'skip': 'Only works from Russia',
187 # video (removed?) only available with list id
188 'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
189 'md5': '091287af5402239a1051c37ec7b92913',
191 'id': '30481095_171201961',
193 'title': 'ТюменцевВВ_09.07.2015',
194 'uploader': 'Anton Ivanov',
196 'upload_date': '20150709',
203 'url': 'https://vk.com/video276849682_170681728',
207 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
208 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a',
210 'upload_date': '20130116',
211 'uploader': "Children's Joy Foundation Inc.",
212 'uploader_id': 'thecjf',
218 'url': 'https://vk.com/video-37468416_456239855',
220 'id': 'k3lz2cmXyRuJQSjGHUv',
222 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f',
223 # TODO: fix test by fixing dailymotion description extraction
224 'description': 'md5:c651358f03c56f1150b555c26d90a0fd',
225 'uploader': 'AniLibria.Tv',
226 'upload_date': '20160914',
227 'uploader_id': 'x1p5vl5',
228 'timestamp': 1473877246,
231 'skip_download': True,
235 # video key is extra_data not url\d+
236 'url': 'http://vk.com/video-110305615_171782105',
237 'md5': 'e13fcda136f99764872e739d13fac1d1',
239 'id': '-110305615_171782105',
241 'title': 'S-Dance, репетиции к The way show',
242 'uploader': 'THE WAY SHOW | 17 апреля',
243 'uploader_id': '-110305615',
244 'timestamp': 1454859345,
245 'upload_date': '20160207',
248 'skip_download': True,
252 # finished live stream, postlive_mp4
253 'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2',
255 'id': '-387766_456242764',
257 'title': 'ИгроМир 2016 День 1 — Игромания Утром',
258 'uploader': 'Игромания',
260 # TODO: use act=show to extract view_count
262 'upload_date': '20160929',
263 'uploader_id': '-387766',
264 'timestamp': 1475137527,
268 # live stream, hls and rtmp links, most likely already finished live
269 # stream by the time you are reading this comment
270 'url': 'https://vk.com/video-140332_456239111',
271 'only_matching': True,
274 # removed video, just testing that we match the pattern
275 'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
276 'only_matching': True,
279 # age restricted video, requires vk account credentials
280 'url': 'https://vk.com/video205387401_164765225',
281 'only_matching': True,
285 'url': 'https://vk.com/video-76116461_171554880',
286 'only_matching': True,
289 'url': 'http://new.vk.com/video205387401_165548505',
290 'only_matching': True,
293 # This video is no longer available, because its author has been blocked.
294 'url': 'https://vk.com/video-10639516_456240611',
295 'only_matching': True,
298 # The video is not available in your region.
299 'url': 'https://vk.com/video-51812607_171445436',
300 'only_matching': True,
303 def _real_extract(self
, url
):
304 mobj
= re
.match(self
._VALID
_URL
, url
)
305 video_id
= mobj
.group('videoid')
308 info_url
= 'https://vk.com/al_video.php?act=show_inline&al=1&video=' + video_id
309 # Some videos (removed?) can only be downloaded with list id specified
310 list_id
= mobj
.group('list_id')
312 info_url
+= '&list=%s' % list_id
314 info_url
= 'http://vk.com/video_ext.php?' + mobj
.group('embed_query')
315 video_id
= '%s_%s' % (mobj
.group('oid'), mobj
.group('id'))
317 info_page
= self
._download
_webpage
(info_url
, video_id
)
319 error_message
= self
._html
_search
_regex
(
320 [r
'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
321 r
'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
322 info_page
, 'error message', default
=None)
324 raise ExtractorError(error_message
, expected
=True)
326 if re
.search(r
'<!>/login\.php\?.*\bact=security_check', info_page
):
327 raise ExtractorError(
328 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
331 ERROR_COPYRIGHT
= 'Video %s has been removed from public access due to rightholder complaint.'
334 r
'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
337 r
'>The video .*? was removed from public access by request of the copyright holder.<':
340 r
'<!>Please log in or <':
341 'Video %s is only available for registered users, '
342 'use --username and --password options to provide account credentials.',
345 'Video %s does not exist.',
347 r
'<!>Видео временно недоступно':
348 'Video %s is temporarily unavailable.',
351 'Access denied to video %s.',
353 r
'<!>Видеозапись недоступна, так как её автор был заблокирован.':
354 'Video %s is no longer available, because its author has been blocked.',
356 r
'<!>This video is no longer available, because its author has been blocked.':
357 'Video %s is no longer available, because its author has been blocked.',
359 r
'<!>This video is no longer available, because it has been deleted.':
360 'Video %s is no longer available, because it has been deleted.',
362 r
'<!>The video .+? is not available in your region.':
363 'Video %s is not available in your region.',
366 for error_re
, error_msg
in ERRORS
.items():
367 if re
.search(error_re
, info_page
):
368 raise ExtractorError(error_msg
% video_id
, expected
=True)
370 youtube_url
= YoutubeIE
._extract
_url
(info_page
)
372 return self
.url_result(youtube_url
, ie
=YoutubeIE
.ie_key())
374 vimeo_url
= VimeoIE
._extract
_url
(url
, info_page
)
375 if vimeo_url
is not None:
376 return self
.url_result(vimeo_url
)
378 pladform_url
= PladformIE
._extract
_url
(info_page
)
380 return self
.url_result(pladform_url
)
382 m_rutube
= re
.search(
383 r
'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page
)
384 if m_rutube
is not None:
385 rutube_url
= self
._proto
_relative
_url
(
386 m_rutube
.group(1).replace('\\', ''))
387 return self
.url_result(rutube_url
)
389 dailymotion_urls
= DailymotionIE
._extract
_urls
(info_page
)
391 return self
.url_result(dailymotion_urls
[0], DailymotionIE
.ie_key())
393 m_opts
= re
.search(r
'(?s)var\s+opts\s*=\s*({.+?});', info_page
)
395 m_opts_url
= re
.search(r
"url\s*:\s*'((?!/\b)[^']+)", m_opts
.group(1))
397 opts_url
= m_opts_url
.group(1)
398 if opts_url
.startswith('//'):
399 opts_url
= 'http:' + opts_url
400 return self
.url_result(opts_url
)
402 # vars does not look to be served anymore since 24.10.2016
403 data
= self
._parse
_json
(
405 r
'var\s+vars\s*=\s*({.+?});', info_page
, 'vars', default
='{}'),
406 video_id
, fatal
=False)
408 # <!json> is served instead
410 data
= self
._parse
_json
(
412 [r
'<!json>\s*({.+?})\s*<!>', r
'<!json>\s*({.+})'],
413 info_page
, 'json', default
='{}'),
416 data
= data
['player']['params'][0]
419 data
= self
._parse
_json
(
421 r
'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', info_page
,
423 video_id
)['params'][0]
425 title
= unescapeHTML(data
['md_title'])
428 # 3 = post live (finished live)
429 is_live
= data
.get('live') == 2
431 title
= self
._live
_title
(title
)
433 timestamp
= unified_timestamp(self
._html
_search
_regex
(
434 r
'class=["\']mv_info_date
[^
>]+>([^
<]+)(?
:<|
from)', info_page,
435 'upload date
', default=None)) or int_or_none(data.get('date
'))
437 view_count = str_to_int(self._search_regex(
438 r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)',
439 info_page, 'view count', default=None))
442 for format_id, format_url in data.items():
443 format_url = url_or_none(format_url)
444 if not format_url or not format_url.startswith(('http', '//', 'rtmp')):
446 if (format_id.startswith(('url', 'cache')) or
447 format_id in ('extra_data', 'live_mp4', 'postlive_mp4')):
448 height = int_or_none(self._search_regex(
449 r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
451 'format_id': format_id,
455 elif format_id == 'hls':
456 formats.extend(self._extract_m3u8_formats(
457 format_url, video_id, 'mp4', 'm3u8_native',
458 m3u8_id=format_id, fatal=False, live=is_live))
459 elif format_id == 'rtmp':
461 'format_id': format_id,
465 self._sort_formats(formats)
471 'thumbnail': data.get('jpg'),
472 'uploader': data.get('md_author'),
473 'uploader_id': str_or_none(data.get('author_id')),
474 'duration': data.get('duration'),
475 'timestamp': timestamp,
476 'view_count': view_count,
477 'like_count': int_or_none(data.get('liked')),
478 'dislike_count': int_or_none(data.get('nolikes')),
483 class VKUserVideosIE(VKBaseIE):
484 IE_NAME = 'vk:uservideos'
485 IE_DESC = "VK
- User
's Videos"
486 _VALID_URL = r'https?
://(?
:(?
:m|new
)\
.)?vk\
.com
/videos(?P
<id>-?
[0-9]+)(?
!\?.*\bz
=video
)(?
:[/?
#&]|$)'
487 _TEMPLATE_URL
= 'https://vk.com/videos'
489 'url': 'http://vk.com/videos205387401',
492 'title': "Tom Cruise's Videos",
494 'playlist_mincount': 4,
496 'url': 'http://vk.com/videos-77521',
497 'only_matching': True,
499 'url': 'http://vk.com/videos-97664626?section=all',
500 'only_matching': True,
502 'url': 'http://m.vk.com/videos205387401',
503 'only_matching': True,
505 'url': 'http://new.vk.com/videos205387401',
506 'only_matching': True,
509 def _real_extract(self
, url
):
510 page_id
= self
._match
_id
(url
)
512 webpage
= self
._download
_webpage
(url
, page_id
)
516 'http://vk.com/video' + video_id
, 'VK', video_id
=video_id
)
517 for video_id
in orderedSet(re
.findall(r
'href="/video(-?[0-9_]+)"', webpage
))]
519 title
= unescapeHTML(self
._search
_regex
(
520 r
'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
521 webpage
, 'title', default
=page_id
))
523 return self
.playlist_result(entries
, page_id
, title
)
526 class VKWallPostIE(VKBaseIE
):
527 IE_NAME
= 'vk:wallpost'
528 _VALID_URL
= r
'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
530 # public page URL, audio playlist
531 'url': 'https://vk.com/bs.official?w=wall-23538238_35',
534 'title': 'Black Shadow - Wall post 23538238_35',
535 'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
538 'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
540 'id': '135220665_111806521',
542 'title': 'Black Shadow - Слепое Верование',
544 'uploader': 'Black Shadow',
545 'artist': 'Black Shadow',
546 'track': 'Слепое Верование',
549 'md5': '4cc7e804579122b17ea95af7834c9233',
551 'id': '135220665_111802303',
553 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
555 'uploader': 'Black Shadow',
556 'artist': 'Black Shadow',
557 'track': 'Война - Негасимое Бездны Пламя!',
560 'skip_download': True,
566 'skip': 'Requires vk account credentials',
568 # single YouTube embed, no leading -
569 'url': 'https://vk.com/wall85155021_6319',
571 'id': '85155021_6319',
572 'title': 'Sergey Gorbunov - Wall post 85155021_6319',
578 'skip': 'Requires vk account credentials',
581 'url': 'https://vk.com/wall-23538238_35',
582 'only_matching': True,
584 # mobile wall page URL
585 'url': 'https://m.vk.com/wall-23538238_35',
586 'only_matching': True,
589 def _real_extract(self
, url
):
590 post_id
= self
._match
_id
(url
)
592 wall_url
= 'https://vk.com/wall%s' % post_id
594 post_id
= remove_start(post_id
, '-')
596 webpage
= self
._download
_webpage
(wall_url
, post_id
)
598 error
= self
._html
_search
_regex
(
599 r
'>Error</div>\s*<div[^>]+class=["\']body
["\'][^>]*>([^<]+)',
600 webpage, 'error', default=None)
602 raise ExtractorError('VK said: %s' % error, expected=True)
604 description = clean_html(get_element_by_class('wall_post_text', webpage))
605 uploader = clean_html(get_element_by_class('author', webpage))
606 thumbnail = self._og_search_thumbnail(webpage)
610 audio_ids = re.findall(r'data-full-id=["\'](\d
+_\d
+)', webpage)
612 al_audio = self._download_webpage(
613 'https
://vk
.com
/al_audio
.php
', post_id,
614 note='Downloading audio info
', fatal=False,
615 data=urlencode_postdata({
616 'act
': 'reload_audio
',
618 'ids
': ','.join(audio_ids)
621 Audio = collections.namedtuple(
622 'Audio
', ['id', 'user_id
', 'url
', 'track
', 'artist
', 'duration
'])
623 audios = self._parse_json(
625 r'<!json
>(.+?
)<!>', al_audio, 'audios
', default='[]'),
626 post_id, fatal=False, transform_source=unescapeHTML)
627 if isinstance(audios, list):
629 a = Audio._make(audio[:6])
631 'id': '%s_%s' % (a.user_id, a.id),
633 'title
': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id,
634 'thumbnail
': thumbnail,
635 'duration
': a.duration,
636 'uploader
': uploader,
641 for video in re.finditer(
642 r'<a
[^
>]+href
=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
643 entries.append(self.url_result(
644 compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
646 title = 'Wall post %s' % post_id
648 return self.playlist_result(
649 orderedSet(entries), post_id,
650 '%s - %s' % (uploader, title) if uploader else title,