diff options
Diffstat (limited to 'hypervideo_dl/extractor/vk.py')
-rw-r--r-- | hypervideo_dl/extractor/vk.py | 140 |
1 files changed, 92 insertions, 48 deletions
diff --git a/hypervideo_dl/extractor/vk.py b/hypervideo_dl/extractor/vk.py index cbc3159..347aa38 100644 --- a/hypervideo_dl/extractor/vk.py +++ b/hypervideo_dl/extractor/vk.py @@ -1,14 +1,17 @@ -# coding: utf-8 -from __future__ import unicode_literals - import collections +import hashlib import re from .common import InfoExtractor +from .dailymotion import DailymotionIE +from .odnoklassniki import OdnoklassnikiIE +from .pladform import PladformIE +from .vimeo import VimeoIE +from .youtube import YoutubeIE from ..compat import compat_urlparse from ..utils import ( - clean_html, ExtractorError, + clean_html, get_element_by_class, int_or_none, orderedSet, @@ -16,19 +19,29 @@ from ..utils import ( str_to_int, unescapeHTML, unified_timestamp, + update_url_query, 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 class VKBaseIE(InfoExtractor): _NETRC_MACHINE = 'vk' + def _download_webpage_handle(self, url_or_request, video_id, *args, fatal=True, **kwargs): + response = super()._download_webpage_handle(url_or_request, video_id, *args, fatal=fatal, **kwargs) + challenge_url, cookie = response[1].geturl() if response else '', None + if challenge_url.startswith('https://vk.com/429.html?'): + cookie = self._get_cookies(challenge_url).get('hash429') + if not cookie: + return response + + hash429 = hashlib.md5(cookie.value.encode('ascii')).hexdigest() + self._request_webpage( + update_url_query(challenge_url, {'key': hash429}), video_id, fatal=fatal, + note='Resolving WAF challenge', errnote='Failed to bypass WAF challenge') + return super()._download_webpage_handle(url_or_request, video_id, *args, fatal=True, **kwargs) + def _perform_login(self, username, password): login_page, url_handle = self._download_webpage_handle( 'https://vk.com', None, 'Downloading login page') @@ -54,11 +67,14 @@ class VKBaseIE(InfoExtractor): 'Unable to login, incorrect username and/or password', expected=True) def _download_payload(self, path, video_id, data, fatal=True): + endpoint = f'https://vk.com/{path}.php' 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'] + endpoint, video_id, data=urlencode_postdata(data), fatal=fatal, + headers={ + 'Referer': endpoint, + 'X-Requested-With': 'XMLHttpRequest', + })['payload'] if code == '3': self.raise_login_required() elif code == '8': @@ -69,6 +85,7 @@ class VKBaseIE(InfoExtractor): class VKIE(VKBaseIE): IE_NAME = 'vk' IE_DESC = 'VK' + _EMBED_REGEX = [r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1'] _VALID_URL = r'''(?x) https?:// (?: @@ -84,20 +101,25 @@ class VKIE(VKBaseIE): (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>([\da-f]+)|(ln-[\da-zA-Z]+)))? ) ''' + # https://help.sibnet.ru/?sibnet_video_embed + _EMBED_REGEX = [r'<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//video\.sibnet\.ru/shell\.php\?.*?\bvideoid=\d+.*?)\1'] _TESTS = [ { 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521', - 'md5': '7babad3b85ea2e91948005b1b8b0cb84', 'info_dict': { 'id': '-77521_162222515', 'ext': 'mp4', 'title': 'ProtivoGunz - Хуёвая песня', 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*', - 'uploader_id': '-77521', + 'uploader_id': '39545378', 'duration': 195, 'timestamp': 1329049880, 'upload_date': '20120212', + 'comment_count': int, + 'like_count': int, + 'thumbnail': r're:https?://.+\.jpg$', }, + 'params': {'skip_download': 'm3u8'}, }, { 'url': 'http://vk.com/video205387401_165548505', @@ -110,12 +132,14 @@ class VKIE(VKBaseIE): 'duration': 9, 'timestamp': 1374364108, 'upload_date': '20130720', + 'comment_count': int, + 'like_count': int, + 'thumbnail': r're:https?://.+\.jpg$', } }, { 'note': 'Embedded video', 'url': 'https://vk.com/video_ext.php?oid=-77521&id=162222515&hash=87b046504ccd8bfa', - 'md5': '7babad3b85ea2e91948005b1b8b0cb84', 'info_dict': { 'id': '-77521_162222515', 'ext': 'mp4', @@ -124,8 +148,10 @@ class VKIE(VKBaseIE): 'duration': 195, 'upload_date': '20120212', 'timestamp': 1329049880, - 'uploader_id': '-77521', + 'uploader_id': '39545378', + 'thumbnail': r're:https?://.+\.jpg$', }, + 'params': {'skip_download': 'm3u8'}, }, { # VIDEO NOW REMOVED @@ -179,8 +205,13 @@ class VKIE(VKBaseIE): 'ext': 'mp4', 'title': '8 серия (озвучка)', 'duration': 8383, + 'comment_count': int, + 'uploader': 'Dizi2021', + 'like_count': int, + 'timestamp': 1640162189, 'upload_date': '20211222', - 'view_count': int, + 'uploader_id': '-93049196', + 'thumbnail': r're:https?://.+\.jpg$', }, }, { @@ -207,10 +238,23 @@ class VKIE(VKBaseIE): 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate", 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a', 'duration': 178, - 'upload_date': '20130116', + 'upload_date': '20130117', 'uploader': "Children's Joy Foundation Inc.", 'uploader_id': 'thecjf', 'view_count': int, + 'channel_id': 'UCgzCNQ11TmR9V97ECnhi3gw', + 'availability': 'public', + 'like_count': int, + 'live_status': 'not_live', + 'playable_in_embed': True, + 'channel': 'Children\'s Joy Foundation Inc.', + 'uploader_url': 'http://www.youtube.com/user/thecjf', + 'thumbnail': r're:https?://.+\.jpg$', + 'tags': 'count:27', + 'start_time': 0.0, + 'categories': ['Nonprofits & Activism'], + 'channel_url': 'https://www.youtube.com/channel/UCgzCNQ11TmR9V97ECnhi3gw', + 'age_limit': 0, }, }, { @@ -226,9 +270,7 @@ class VKIE(VKBaseIE): 'uploader_id': 'x1p5vl5', 'timestamp': 1473877246, }, - 'params': { - 'skip_download': True, - }, + 'skip': 'Removed' }, { # video key is extra_data not url\d+ @@ -243,9 +285,7 @@ class VKIE(VKBaseIE): 'timestamp': 1454859345, 'upload_date': '20160207', }, - 'params': { - 'skip_download': True, - }, + 'skip': 'Removed', }, { # finished live stream, postlive_mp4 @@ -256,11 +296,12 @@ class VKIE(VKBaseIE): 'title': 'ИгроМир 2016 День 1 — Игромания Утром', 'uploader': 'Игромания', 'duration': 5239, - # TODO: use act=show to extract view_count - # 'view_count': int, 'upload_date': '20160929', 'uploader_id': '-387766', 'timestamp': 1475137527, + 'thumbnail': r're:https?://.+\.jpg$', + 'comment_count': int, + 'like_count': int, }, 'params': { 'skip_download': True, @@ -306,13 +347,6 @@ class VKIE(VKBaseIE): 'only_matching': True, }] - @staticmethod - def _extract_sibnet_urls(webpage): - # https://help.sibnet.ru/?sibnet_video_embed - return [unescapeHTML(mobj.group('url')) for mobj in re.finditer( - r'<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//video\.sibnet\.ru/shell\.php\?.*?\bvideoid=\d+.*?)\1', - webpage)] - def _real_extract(self, url): mobj = self._match_valid_url(url) video_id = mobj.group('videoid') @@ -320,7 +354,7 @@ class VKIE(VKBaseIE): mv_data = {} if video_id: data = { - 'act': 'show_inline', + 'act': 'show', 'video': video_id, } # Some videos (removed?) can only be downloaded with list id specified @@ -413,17 +447,17 @@ class VKIE(VKBaseIE): m_rutube.group(1).replace('\\', '')) return self.url_result(rutube_url) - dailymotion_urls = DailymotionIE._extract_urls(info_page) - if dailymotion_urls: - return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key()) + dailymotion_url = next(DailymotionIE._extract_embed_urls(url, info_page), None) + if dailymotion_url: + return self.url_result(dailymotion_url, DailymotionIE.ie_key()) odnoklassniki_url = OdnoklassnikiIE._extract_url(info_page) if odnoklassniki_url: return self.url_result(odnoklassniki_url, OdnoklassnikiIE.ie_key()) - sibnet_urls = self._extract_sibnet_urls(info_page) - if sibnet_urls: - return self.url_result(sibnet_urls[0]) + sibnet_url = next(self._extract_embed_urls(url, info_page), None) + if sibnet_url: + return self.url_result(sibnet_url) m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page) if m_opts: @@ -473,7 +507,6 @@ class VKIE(VKBaseIE): 'url': format_url, 'ext': 'flv', }) - self._sort_formats(formats) subtitles = {} for sub in data.get('subs') or {}: @@ -502,7 +535,7 @@ class VKIE(VKBaseIE): class VKUserVideosIE(VKBaseIE): IE_NAME = 'vk:uservideos' IE_DESC = "VK - User's Videos" - _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/video/@(?P<id>[^?$#/&]+)(?!\?.*\bz=video)(?:[/?#&](?:.*?\bsection=(?P<section>\w+))?|$)' + _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/video/(?:playlist/)?(?P<id>[^?$#/&]+)(?!\?.*\bz=video)(?:[/?#&](?:.*?\bsection=(?P<section>\w+))?|$)' _TEMPLATE_URL = 'https://vk.com/videos' _TESTS = [{ 'url': 'https://vk.com/video/@mobidevices', @@ -516,6 +549,13 @@ class VKUserVideosIE(VKBaseIE): 'id': '-17892518_uploaded', }, 'playlist_mincount': 182, + }, { + 'url': 'https://vk.com/video/playlist/-174476437_2', + 'info_dict': { + 'id': '-174476437_2', + 'title': 'Анонсы' + }, + 'playlist_mincount': 108, }] _VIDEO = collections.namedtuple('Video', ['owner_id', 'id']) @@ -550,11 +590,19 @@ class VKUserVideosIE(VKBaseIE): def _real_extract(self, url): u_id, section = self._match_valid_url(url).groups() webpage = self._download_webpage(url, u_id) - page_id = self._search_regex(r'data-owner-id\s?=\s?"([^"]+)"', webpage, 'page_id') + + if u_id.startswith('@'): + page_id = self._search_regex(r'data-owner-id\s?=\s?"([^"]+)"', webpage, 'page_id') + elif '_' in u_id: + page_id, section = u_id.split('_', 1) + else: + raise ExtractorError('Invalid URL', expected=True) + if not section: section = 'all' - return self.playlist_result(self._entries(page_id, section), '%s_%s' % (page_id, section)) + playlist_title = clean_html(get_element_by_class('VideoInfoPanel__title', webpage)) + return self.playlist_result(self._entries(page_id, section), '%s_%s' % (page_id, section), playlist_title) class VKWallPostIE(VKBaseIE): @@ -593,7 +641,6 @@ class VKWallPostIE(VKBaseIE): }], 'params': { 'skip_download': True, - 'usenetrc': True, }, 'skip': 'Requires vk account credentials', }, { @@ -604,9 +651,6 @@ class VKWallPostIE(VKBaseIE): 'title': 'Сергей Горбунов - Wall post 85155021_6319', }, 'playlist_count': 1, - 'params': { - 'usenetrc': True, - }, 'skip': 'Requires vk account credentials', }, { # wall page URL |