diff options
Diffstat (limited to 'hypervideo_dl/extractor/vk.py')
-rw-r--r-- | hypervideo_dl/extractor/vk.py | 118 |
1 files changed, 63 insertions, 55 deletions
diff --git a/hypervideo_dl/extractor/vk.py b/hypervideo_dl/extractor/vk.py index d8a9b9a..cbc3159 100644 --- a/hypervideo_dl/extractor/vk.py +++ b/hypervideo_dl/extractor/vk.py @@ -2,7 +2,6 @@ from __future__ import unicode_literals import collections -import functools import re from .common import InfoExtractor @@ -12,7 +11,6 @@ from ..utils import ( ExtractorError, get_element_by_class, int_or_none, - OnDemandPagedList, orderedSet, str_or_none, str_to_int, @@ -31,11 +29,7 @@ from .youtube import YoutubeIE class VKBaseIE(InfoExtractor): _NETRC_MACHINE = 'vk' - def _login(self): - username, password = self._get_login_info() - if username is None: - return - + def _perform_login(self, username, password): login_page, url_handle = self._download_webpage_handle( 'https://vk.com', None, 'Downloading login page') @@ -51,7 +45,7 @@ class VKBaseIE(InfoExtractor): self._apply_first_set_cookie_header(url_handle, 'remixlhk') login_page = self._download_webpage( - 'https://login.vk.com/?act=login', None, + 'https://vk.com/login', None, note='Logging in', data=urlencode_postdata(login_form)) @@ -59,9 +53,6 @@ class VKBaseIE(InfoExtractor): raise ExtractorError( 'Unable to login, incorrect username and/or password', expected=True) - def _real_initialize(self): - self._login() - def _download_payload(self, path, video_id, data, fatal=True): data['al'] = 1 code, payload = self._download_json( @@ -87,10 +78,10 @@ class VKIE(VKBaseIE): ) ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)| (?: - (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video| + (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?(?:video|clip)| (?:www\.)?daxab.com/embed/ ) - (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))? + (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>([\da-f]+)|(ln-[\da-zA-Z]+)))? ) ''' _TESTS = [ @@ -182,6 +173,17 @@ class VKIE(VKBaseIE): 'skip': 'Removed', }, { + 'url': 'https://vk.com/video-93049196_456239755?list=ln-cBjJ7S4jYYx3ADnmDT', + 'info_dict': { + 'id': '-93049196_456239755', + 'ext': 'mp4', + 'title': '8 серия (озвучка)', + 'duration': 8383, + 'upload_date': '20211222', + 'view_count': int, + }, + }, + { # video (removed?) only available with list id 'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4', 'md5': '091287af5402239a1051c37ec7b92913', @@ -298,6 +300,10 @@ class VKIE(VKBaseIE): # The video is not available in your region. 'url': 'https://vk.com/video-51812607_171445436', 'only_matching': True, + }, + { + 'url': 'https://vk.com/clip30014565_456240946', + 'only_matching': True, }] @staticmethod @@ -434,8 +440,6 @@ class VKIE(VKBaseIE): # 2 = live # 3 = post live (finished live) is_live = data.get('live') == 2 - if is_live: - title = self._live_title(title) timestamp = unified_timestamp(self._html_search_regex( r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page, @@ -471,6 +475,13 @@ class VKIE(VKBaseIE): }) self._sort_formats(formats) + subtitles = {} + for sub in data.get('subs') or {}: + subtitles.setdefault(sub.get('lang', 'en'), []).append({ + 'ext': sub.get('title', '.srt').split('.')[-1], + 'url': url_or_none(sub.get('url')), + }) + return { 'id': video_id, 'formats': formats, @@ -484,69 +495,66 @@ class VKIE(VKBaseIE): 'like_count': int_or_none(mv_data.get('likes')), 'comment_count': int_or_none(mv_data.get('commcount')), 'is_live': is_live, + 'subtitles': subtitles, } class VKUserVideosIE(VKBaseIE): IE_NAME = 'vk:uservideos' IE_DESC = "VK - User's Videos" - _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&](?:.*?\bsection=(?P<section>\w+))?|$)' + _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/video/@(?P<id>[^?$#/&]+)(?!\?.*\bz=video)(?:[/?#&](?:.*?\bsection=(?P<section>\w+))?|$)' _TEMPLATE_URL = 'https://vk.com/videos' _TESTS = [{ - 'url': 'https://vk.com/videos-767561', + 'url': 'https://vk.com/video/@mobidevices', 'info_dict': { - 'id': '-767561_all', + 'id': '-17892518_all', }, - 'playlist_mincount': 1150, + 'playlist_mincount': 1355, }, { - 'url': 'https://vk.com/videos-767561?section=uploaded', + 'url': 'https://vk.com/video/@mobidevices?section=uploaded', 'info_dict': { - 'id': '-767561_uploaded', + 'id': '-17892518_uploaded', }, - 'playlist_mincount': 425, - }, { - 'url': 'http://vk.com/videos205387401', - 'only_matching': True, - }, { - 'url': 'http://vk.com/videos-77521', - 'only_matching': True, - }, { - 'url': 'http://vk.com/videos-97664626?section=all', - 'only_matching': True, - }, { - 'url': 'http://m.vk.com/videos205387401', - 'only_matching': True, - }, { - 'url': 'http://new.vk.com/videos205387401', - 'only_matching': True, + 'playlist_mincount': 182, }] - _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, { + def _entries(self, page_id, section): + video_list_json = self._download_payload('al_video', page_id, { 'act': 'load_videos_silent', - 'offset': page * self._PAGE_SIZE, + 'offset': 0, '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) + })[0][section] + count = video_list_json['count'] + total = video_list_json['total'] + video_list = video_list_json['list'] + + while True: + for video in video_list: + 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) + if count >= total: + break + video_list_json = self._download_payload('al_video', page_id, { + 'act': 'load_videos_silent', + 'offset': count, + 'oid': page_id, + 'section': section, + })[0][section] + count += video_list_json['count'] + video_list = video_list_json['list'] def _real_extract(self, url): - page_id, section = self._match_valid_url(url).groups() + 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 not section: section = 'all' - entries = OnDemandPagedList( - functools.partial(self._fetch_page, page_id, section), - self._PAGE_SIZE) - - return self.playlist_result(entries, '%s_%s' % (page_id, section)) + return self.playlist_result(self._entries(page_id, section), '%s_%s' % (page_id, section)) class VKWallPostIE(VKBaseIE): @@ -673,7 +681,7 @@ class VKWallPostIE(VKBaseIE): 'artist': performer, 'track': title, 'ext': 'mp4', - 'protocol': 'm3u8', + 'protocol': 'm3u8_native', }) for video in re.finditer( |