X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/a4409a39ae30315a4f0c7cc8eb7c8f9a937225bc..e22397a1c5329e12f3719961c9456e378f94dbb1:/youtube_dl/extractor/vk.py diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index b50d4f1..00ec006 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -2,27 +2,27 @@ from __future__ import unicode_literals import collections +import functools import re -import sys from .common import InfoExtractor -from ..compat import ( - compat_str, - compat_urlparse, -) +from ..compat import compat_urlparse from ..utils import ( clean_html, ExtractorError, get_element_by_class, int_or_none, + OnDemandPagedList, orderedSet, - remove_start, + str_or_none, str_to_int, unescapeHTML, unified_timestamp, + url_or_none, urlencode_postdata, ) from .dailymotion import DailymotionIE +from .odnoklassniki import OdnoklassnikiIE from .pladform import PladformIE from .vimeo import VimeoIE from .youtube import YoutubeIE @@ -32,7 +32,7 @@ class VKBaseIE(InfoExtractor): _NETRC_MACHINE = 'vk' def _login(self): - (username, password) = self._get_login_info() + username, password = self._get_login_info() if username is None: return @@ -46,24 +46,9 @@ class VKBaseIE(InfoExtractor): 'pass': password.encode('cp1251'), }) - # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header - # and expects the first one to be set rather than second (see - # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201). - # As of RFC6265 the newer one cookie should be set into cookie store - # what actually happens. - # We will workaround this VK issue by resetting the remixlhk cookie to - # the first one manually. - for header, cookies in url_handle.headers.items(): - if header.lower() != 'set-cookie': - continue - if sys.version_info[0] >= 3: - cookies = cookies.encode('iso-8859-1') - cookies = cookies.decode('utf-8') - remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies) - if remixlhk: - value, domain = remixlhk.groups() - self._set_cookie(domain, 'remixlhk', value) - break + # vk serves two same remixlhk cookies in Set-Cookie header and expects + # first one to be actually set + self._apply_first_set_cookie_header(url_handle, 'remixlhk') login_page = self._download_webpage( 'https://login.vk.com/?act=login', None, @@ -77,6 +62,18 @@ class VKBaseIE(InfoExtractor): def _real_initialize(self): self._login() + def _download_payload(self, path, video_id, data, fatal=True): + data['al'] = 1 + code, payload = self._download_json( + 'https://vk.com/%s.php' % path, video_id, + data=urlencode_postdata(data), fatal=fatal, + headers={'X-Requested-With': 'XMLHttpRequest'})['payload'] + if code == '3': + self.raise_login_required() + elif code == '8': + raise ExtractorError(clean_html(payload[0][1:-1]), expected=True) + return payload + class VKIE(VKBaseIE): IE_NAME = 'vk' @@ -101,44 +98,43 @@ class VKIE(VKBaseIE): 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521', 'md5': '7babad3b85ea2e91948005b1b8b0cb84', 'info_dict': { - 'id': '162222515', + 'id': '-77521_162222515', 'ext': 'mp4', 'title': 'ProtivoGunz - Хуёвая песня', 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*', + 'uploader_id': '-77521', 'duration': 195, - 'timestamp': 1329060660, + 'timestamp': 1329049880, 'upload_date': '20120212', - 'view_count': int, }, }, { 'url': 'http://vk.com/video205387401_165548505', - 'md5': '6c0aeb2e90396ba97035b9cbde548700', 'info_dict': { - 'id': '165548505', + 'id': '205387401_165548505', 'ext': 'mp4', - 'uploader': 'Tom Cruise', 'title': 'No name', + 'uploader': 'Tom Cruise', + 'uploader_id': '205387401', 'duration': 9, - 'timestamp': 1374374880, - 'upload_date': '20130721', - 'view_count': int, + 'timestamp': 1374364108, + 'upload_date': '20130720', } }, { 'note': 'Embedded video', - 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1', - 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a', + 'url': 'https://vk.com/video_ext.php?oid=-77521&id=162222515&hash=87b046504ccd8bfa', + 'md5': '7babad3b85ea2e91948005b1b8b0cb84', 'info_dict': { - 'id': '162925554', + 'id': '-77521_162222515', 'ext': 'mp4', - 'uploader': 'Vladimir Gavrin', - 'title': 'Lin Dan', - 'duration': 101, - 'upload_date': '20120730', - 'view_count': int, + 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*', + 'title': 'ProtivoGunz - Хуёвая песня', + 'duration': 195, + 'upload_date': '20120212', + 'timestamp': 1329049880, + 'uploader_id': '-77521', }, - 'skip': 'This video has been removed from public access.', }, { # VIDEO NOW REMOVED @@ -147,7 +143,7 @@ class VKIE(VKBaseIE): 'md5': 'a590bcaf3d543576c9bd162812387666', 'note': 'Only available for registered users', 'info_dict': { - 'id': '164049491', + 'id': '-8871596_164049491', 'ext': 'mp4', 'uploader': 'Триллеры', 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]', @@ -155,18 +151,19 @@ class VKIE(VKBaseIE): 'upload_date': '20121218', 'view_count': int, }, - 'skip': 'Requires vk account credentials', + 'skip': 'Removed', }, { 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d', - 'md5': '4d7a5ef8cf114dfa09577e57b2993202', 'info_dict': { - 'id': '168067957', + 'id': '-43215063_168067957', 'ext': 'mp4', - 'uploader': 'Киномания - лучшее из мира кино', + 'uploader': 'Bro Mazter', 'title': ' ', 'duration': 7291, 'upload_date': '20140328', + 'uploader_id': '223413403', + 'timestamp': 1396018030, }, 'skip': 'Requires vk account credentials', }, @@ -175,21 +172,21 @@ class VKIE(VKBaseIE): 'md5': '0c45586baa71b7cb1d0784ee3f4e00a6', 'note': 'ivi.ru embed', 'info_dict': { - 'id': '60690', + 'id': '-43215063_169084319', 'ext': 'mp4', 'title': 'Книга Илая', 'duration': 6771, 'upload_date': '20140626', 'view_count': int, }, - 'skip': 'Only works from Russia', + 'skip': 'Removed', }, { # video (removed?) only available with list id 'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4', 'md5': '091287af5402239a1051c37ec7b92913', 'info_dict': { - 'id': '171201961', + 'id': '30481095_171201961', 'ext': 'mp4', 'title': 'ТюменцевВВ_09.07.2015', 'uploader': 'Anton Ivanov', @@ -204,12 +201,12 @@ class VKIE(VKBaseIE): 'url': 'https://vk.com/video276849682_170681728', 'info_dict': { 'id': 'V3K4mi0SYkc', - 'ext': 'webm', + 'ext': 'mp4', 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate", - 'description': 'md5:d9903938abdc74c738af77f527ca0596', + 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a', 'duration': 178, 'upload_date': '20130116', - 'uploader': "Children's Joy Foundation", + 'uploader': "Children's Joy Foundation Inc.", 'uploader_id': 'thecjf', 'view_count': int, }, @@ -221,7 +218,7 @@ class VKIE(VKBaseIE): 'id': 'k3lz2cmXyRuJQSjGHUv', 'ext': 'mp4', 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f', - 'description': 'md5:c651358f03c56f1150b555c26d90a0fd', + 'description': 'md5:424b8e88cc873217f520e582ba28bb36', 'uploader': 'AniLibria.Tv', 'upload_date': '20160914', 'uploader_id': 'x1p5vl5', @@ -236,26 +233,35 @@ class VKIE(VKBaseIE): 'url': 'http://vk.com/video-110305615_171782105', 'md5': 'e13fcda136f99764872e739d13fac1d1', 'info_dict': { - 'id': '171782105', + 'id': '-110305615_171782105', 'ext': 'mp4', 'title': 'S-Dance, репетиции к The way show', 'uploader': 'THE WAY SHOW | 17 апреля', - 'timestamp': 1454870100, + 'uploader_id': '-110305615', + 'timestamp': 1454859345, 'upload_date': '20160207', - 'view_count': int, + }, + 'params': { + 'skip_download': True, }, }, { # finished live stream, postlive_mp4 'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2', - 'md5': '90d22d051fccbbe9becfccc615be6791', 'info_dict': { - 'id': '456242764', + 'id': '-387766_456242764', 'ext': 'mp4', - 'title': 'ИгроМир 2016 — день 1', + 'title': 'ИгроМир 2016 День 1 — Игромания Утром', 'uploader': 'Игромания', 'duration': 5239, - 'view_count': int, + # TODO: use act=show to extract view_count + # 'view_count': int, + 'upload_date': '20160929', + 'uploader_id': '-387766', + 'timestamp': 1475137527, + }, + 'params': { + 'skip_download': True, }, }, { @@ -287,81 +293,105 @@ class VKIE(VKBaseIE): # This video is no longer available, because its author has been blocked. 'url': 'https://vk.com/video-10639516_456240611', 'only_matching': True, - } - ] + }, + { + # The video is not available in your region. + 'url': 'https://vk.com/video-51812607_171445436', + 'only_matching': True, + }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('videoid') + mv_data = {} if video_id: - info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id + data = { + 'act': 'show_inline', + 'video': video_id, + } # Some videos (removed?) can only be downloaded with list id specified list_id = mobj.group('list_id') if list_id: - info_url += '&list=%s' % list_id + data['list'] = list_id + + payload = self._download_payload('al_video', video_id, data) + info_page = payload[1] + opts = payload[-1] + mv_data = opts.get('mvData') or {} + player = opts.get('player') or {} else: - info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query') video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id')) - info_page = self._download_webpage(info_url, video_id) + info_page = self._download_webpage( + 'http://vk.com/video_ext.php?' + mobj.group('embed_query'), video_id) - error_message = self._html_search_regex( - [r'(?s)]+class="video_layer_message"[^>]*>(.+?)', - r'(?s)]+id="video_ext_msg"[^>]*>(.+?)'], - info_page, 'error message', default=None) - if error_message: - raise ExtractorError(error_message, expected=True) + error_message = self._html_search_regex( + [r'(?s)]+class="video_layer_message"[^>]*>(.+?)', + r'(?s)]+id="video_ext_msg"[^>]*>(.+?)'], + info_page, 'error message', default=None) + if error_message: + raise ExtractorError(error_message, expected=True) - if re.search(r'/login\.php\?.*\bact=security_check', info_page): - raise ExtractorError( - 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.', - expected=True) + if re.search(r'/login\.php\?.*\bact=security_check', info_page): + raise ExtractorError( + 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.', + expected=True) - ERROR_COPYRIGHT = 'Video %s has been removed from public access due to rightholder complaint.' + ERROR_COPYRIGHT = 'Video %s has been removed from public access due to rightholder complaint.' - ERRORS = { - r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<': - ERROR_COPYRIGHT, + ERRORS = { + r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<': + ERROR_COPYRIGHT, - r'>The video .*? was removed from public access by request of the copyright holder.<': - ERROR_COPYRIGHT, + r'>The video .*? was removed from public access by request of the copyright holder.<': + ERROR_COPYRIGHT, - r'Please log in or <': - 'Video %s is only available for registered users, ' - 'use --username and --password options to provide account credentials.', + r'Please log in or <': + 'Video %s is only available for registered users, ' + 'use --username and --password options to provide account credentials.', - r'Unknown error': - 'Video %s does not exist.', + r'Unknown error': + 'Video %s does not exist.', - r'Видео временно недоступно': - 'Video %s is temporarily unavailable.', + r'Видео временно недоступно': + 'Video %s is temporarily unavailable.', - r'Access denied': - 'Access denied to video %s.', + r'Access denied': + 'Access denied to video %s.', - r'Видеозапись недоступна, так как её автор был заблокирован.': - 'Video %s is no longer available, because its author has been blocked.', + r'Видеозапись недоступна, так как её автор был заблокирован.': + 'Video %s is no longer available, because its author has been blocked.', - r'This video is no longer available, because its author has been blocked.': - 'Video %s is no longer available, because its author has been blocked.', - } + r'This video is no longer available, because its author has been blocked.': + 'Video %s is no longer available, because its author has been blocked.', + + r'This video is no longer available, because it has been deleted.': + 'Video %s is no longer available, because it has been deleted.', + + r'The video .+? is not available in your region.': + 'Video %s is not available in your region.', + } + + for error_re, error_msg in ERRORS.items(): + if re.search(error_re, info_page): + raise ExtractorError(error_msg % video_id, expected=True) - for error_re, error_msg in ERRORS.items(): - if re.search(error_re, info_page): - raise ExtractorError(error_msg % video_id, expected=True) + player = self._parse_json(self._search_regex( + r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', + info_page, 'player params'), video_id) youtube_url = YoutubeIE._extract_url(info_page) if youtube_url: - return self.url_result(youtube_url, ie=YoutubeIE.ie_key()) + return self.url_result(youtube_url, YoutubeIE.ie_key()) vimeo_url = VimeoIE._extract_url(url, info_page) if vimeo_url is not None: - return self.url_result(vimeo_url) + return self.url_result(vimeo_url, VimeoIE.ie_key()) pladform_url = PladformIE._extract_url(info_page) if pladform_url: - return self.url_result(pladform_url) + return self.url_result(pladform_url, PladformIE.ie_key()) m_rutube = re.search( r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page) @@ -374,6 +404,10 @@ class VKIE(VKBaseIE): if dailymotion_urls: return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key()) + odnoklassniki_url = OdnoklassnikiIE._extract_url(info_page) + if odnoklassniki_url: + return self.url_result(odnoklassniki_url, OdnoklassnikiIE.ie_key()) + m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page) if m_opts: m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1)) @@ -383,28 +417,7 @@ class VKIE(VKBaseIE): opts_url = 'http:' + opts_url return self.url_result(opts_url) - # vars does not look to be served anymore since 24.10.2016 - data = self._parse_json( - self._search_regex( - r'var\s+vars\s*=\s*({.+?});', info_page, 'vars', default='{}'), - video_id, fatal=False) - - # is served instead - if not data: - data = self._parse_json( - self._search_regex( - r'\s*({.+?})\s*', info_page, 'json', default='{}'), - video_id) - if data: - data = data['player']['params'][0] - - if not data: - data = self._parse_json( - self._search_regex( - r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', info_page, - 'player params'), - video_id)['params'][0] - + data = player['params'][0] title = unescapeHTML(data['md_title']) # 2 = live @@ -415,7 +428,7 @@ class VKIE(VKBaseIE): timestamp = unified_timestamp(self._html_search_regex( r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page, - 'upload date', fatal=False)) + 'upload date', default=None)) or int_or_none(data.get('date')) view_count = str_to_int(self._search_regex( r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)', @@ -423,10 +436,11 @@ class VKIE(VKBaseIE): formats = [] for format_id, format_url in data.items(): - if not isinstance(format_url, compat_str) or not format_url.startswith(('http', '//', 'rtmp')): + format_url = url_or_none(format_url) + if not format_url or not format_url.startswith(('http', '//', 'rtmp')): continue - if (format_id.startswith(('url', 'cache')) or - format_id in ('extra_data', 'live_mp4', 'postlive_mp4')): + if (format_id.startswith(('url', 'cache')) + or format_id in ('extra_data', 'live_mp4', 'postlive_mp4')): height = int_or_none(self._search_regex( r'^(?:url|cache)(\d+)', format_id, 'height', default=None)) formats.append({ @@ -447,14 +461,17 @@ class VKIE(VKBaseIE): self._sort_formats(formats) return { - 'id': compat_str(data.get('vid') or video_id), + 'id': video_id, 'formats': formats, 'title': title, 'thumbnail': data.get('jpg'), 'uploader': data.get('md_author'), - 'duration': data.get('duration'), + 'uploader_id': str_or_none(data.get('author_id') or mv_data.get('authorId')), + 'duration': int_or_none(data.get('duration') or mv_data.get('duration')), 'timestamp': timestamp, 'view_count': view_count, + 'like_count': int_or_none(mv_data.get('likes')), + 'comment_count': int_or_none(mv_data.get('commcount')), 'is_live': is_live, } @@ -462,15 +479,23 @@ class VKIE(VKBaseIE): class VKUserVideosIE(VKBaseIE): IE_NAME = 'vk:uservideos' IE_DESC = "VK - User's Videos" - _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)' + _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&](?:.*?\bsection=(?P
\w+))?|$)' _TEMPLATE_URL = 'https://vk.com/videos' _TESTS = [{ - 'url': 'http://vk.com/videos205387401', + 'url': 'https://vk.com/videos-767561', 'info_dict': { - 'id': '205387401', - 'title': "Tom Cruise's Videos", + 'id': '-767561_all', }, - 'playlist_mincount': 4, + 'playlist_mincount': 1150, + }, { + 'url': 'https://vk.com/videos-767561?section=uploaded', + 'info_dict': { + 'id': '-767561_uploaded', + }, + 'playlist_mincount': 425, + }, { + 'url': 'http://vk.com/videos205387401', + 'only_matching': True, }, { 'url': 'http://vk.com/videos-77521', 'only_matching': True, @@ -484,22 +509,33 @@ class VKUserVideosIE(VKBaseIE): 'url': 'http://new.vk.com/videos205387401', 'only_matching': True, }] + _PAGE_SIZE = 1000 + _VIDEO = collections.namedtuple('Video', ['owner_id', 'id']) + + def _fetch_page(self, page_id, section, page): + l = self._download_payload('al_video', page_id, { + 'act': 'load_videos_silent', + 'offset': page * self._PAGE_SIZE, + 'oid': page_id, + 'section': section, + })[0][section]['list'] + + for video in l: + v = self._VIDEO._make(video[:2]) + video_id = '%d_%d' % (v.owner_id, v.id) + yield self.url_result( + 'http://vk.com/video' + video_id, VKIE.ie_key(), video_id) def _real_extract(self, url): - page_id = self._match_id(url) - - webpage = self._download_webpage(url, page_id) - - entries = [ - self.url_result( - 'http://vk.com/video' + video_id, 'VK', video_id=video_id) - for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))] + page_id, section = re.match(self._VALID_URL, url).groups() + if not section: + section = 'all' - title = unescapeHTML(self._search_regex( - r'\s*([^<]+?)\s+\|\s+\d+\s+videos', - webpage, 'title', default=page_id)) + entries = OnDemandPagedList( + functools.partial(self._fetch_page, page_id, section), + self._PAGE_SIZE) - return self.playlist_result(entries, page_id, title) + return self.playlist_result(entries, '%s_%s' % (page_id, section)) class VKWallPostIE(VKBaseIE): @@ -509,15 +545,15 @@ class VKWallPostIE(VKBaseIE): # public page URL, audio playlist 'url': 'https://vk.com/bs.official?w=wall-23538238_35', 'info_dict': { - 'id': '23538238_35', - 'title': 'Black Shadow - Wall post 23538238_35', + 'id': '-23538238_35', + 'title': 'Black Shadow - Wall post -23538238_35', 'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c', }, 'playlist': [{ 'md5': '5ba93864ec5b85f7ce19a9af4af080f6', 'info_dict': { 'id': '135220665_111806521', - 'ext': 'mp3', + 'ext': 'mp4', 'title': 'Black Shadow - Слепое Верование', 'duration': 370, 'uploader': 'Black Shadow', @@ -528,18 +564,16 @@ class VKWallPostIE(VKBaseIE): 'md5': '4cc7e804579122b17ea95af7834c9233', 'info_dict': { 'id': '135220665_111802303', - 'ext': 'mp3', + 'ext': 'mp4', 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!', 'duration': 423, 'uploader': 'Black Shadow', 'artist': 'Black Shadow', 'track': 'Война - Негасимое Бездны Пламя!', }, - 'params': { - 'skip_download': True, - }, }], 'params': { + 'skip_download': True, 'usenetrc': True, }, 'skip': 'Requires vk account credentials', @@ -548,7 +582,7 @@ class VKWallPostIE(VKBaseIE): 'url': 'https://vk.com/wall85155021_6319', 'info_dict': { 'id': '85155021_6319', - 'title': 'Sergey Gorbunov - Wall post 85155021_6319', + 'title': 'Сергей Горбунов - Wall post 85155021_6319', }, 'playlist_count': 1, 'params': { @@ -564,58 +598,72 @@ class VKWallPostIE(VKBaseIE): 'url': 'https://m.vk.com/wall-23538238_35', 'only_matching': True, }] + _BASE64_CHARS = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMN0PQRSTUVWXYZO123456789+/=' + _AUDIO = collections.namedtuple('Audio', ['id', 'owner_id', 'url', 'title', 'performer', 'duration', 'album_id', 'unk', 'author_link', 'lyrics', 'flags', 'context', 'extra', 'hashes', 'cover_url', 'ads']) + + def _decode(self, enc): + dec = '' + e = n = 0 + for c in enc: + r = self._BASE64_CHARS.index(c) + cond = n % 4 + e = 64 * e + r if cond else r + n += 1 + if cond: + dec += chr(255 & e >> (-2 * n & 6)) + return dec + + def _unmask_url(self, mask_url, vk_id): + if 'audio_api_unavailable' in mask_url: + extra = mask_url.split('?extra=')[1].split('#') + func, base = self._decode(extra[1]).split(chr(11)) + mask_url = list(self._decode(extra[0])) + url_len = len(mask_url) + indexes = [None] * url_len + index = int(base) ^ vk_id + for n in range(url_len - 1, -1, -1): + index = (url_len * (n + 1) ^ index + n) % url_len + indexes[n] = index + for n in range(1, url_len): + c = mask_url[n] + index = indexes[url_len - 1 - n] + mask_url[n] = mask_url[index] + mask_url[index] = c + mask_url = ''.join(mask_url) + return mask_url def _real_extract(self, url): post_id = self._match_id(url) - wall_url = 'https://vk.com/wall%s' % post_id - - post_id = remove_start(post_id, '-') - - webpage = self._download_webpage(wall_url, post_id) - - error = self._html_search_regex( - r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)', - webpage, 'error', default=None) - if error: - raise ExtractorError('VK said: %s' % error, expected=True) + webpage = self._download_payload('wkview', post_id, { + 'act': 'show', + 'w': 'wall' + post_id, + })[1] description = clean_html(get_element_by_class('wall_post_text', webpage)) uploader = clean_html(get_element_by_class('author', webpage)) - thumbnail = self._og_search_thumbnail(webpage) entries = [] - audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage) - if audio_ids: - al_audio = self._download_webpage( - 'https://vk.com/al_audio.php', post_id, - note='Downloading audio info', fatal=False, - data=urlencode_postdata({ - 'act': 'reload_audio', - 'al': '1', - 'ids': ','.join(audio_ids) - })) - if al_audio: - Audio = collections.namedtuple( - 'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration']) - audios = self._parse_json( - self._search_regex( - r'<!json>(.+?)<!>', al_audio, 'audios', default='[]'), - post_id, fatal=False, transform_source=unescapeHTML) - if isinstance(audios, list): - for audio in audios: - a = Audio._make(audio[:6]) - entries.append({ - 'id': '%s_%s' % (a.user_id, a.id), - 'url': a.url, - 'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id, - 'thumbnail': thumbnail, - 'duration': a.duration, - 'uploader': uploader, - 'artist': a.artist, - 'track': a.track, - }) + for audio in re.findall(r'data-audio="([^"]+)', webpage): + audio = self._parse_json(unescapeHTML(audio), post_id) + a = self._AUDIO._make(audio[:16]) + if not a.url: + continue + title = unescapeHTML(a.title) + performer = unescapeHTML(a.performer) + entries.append({ + 'id': '%s_%s' % (a.owner_id, a.id), + 'url': self._unmask_url(a.url, a.ads['vk_id']), + 'title': '%s - %s' % (performer, title) if performer else title, + 'thumbnails': [{'url': c_url} for c_url in a.cover_url.split(',')] if a.cover_url else None, + 'duration': int_or_none(a.duration), + 'uploader': uploader, + 'artist': performer, + 'track': title, + 'ext': 'mp4', + 'protocol': 'm3u8', + }) for video in re.finditer( r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):