X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fvk.py;h=b52d15ac6f0709dce8a60c2f42d690b49a56d391;hb=2f483bc1c389709623117079439708783122b5ec;hp=df43ba86755eca4d25e5c670626550eb4bc88053;hpb=9218a6b4f5292cc3d7761484e6b5dbe1e7a1a998;p=youtube-dl diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index df43ba867..b52d15ac6 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -3,7 +3,6 @@ from __future__ import unicode_literals import collections import re -import json import sys from .common import InfoExtractor @@ -18,21 +17,24 @@ from ..utils import ( int_or_none, orderedSet, remove_start, + str_or_none, str_to_int, unescapeHTML, unified_timestamp, + url_or_none, urlencode_postdata, ) from .dailymotion import DailymotionIE from .pladform import PladformIE from .vimeo import VimeoIE +from .youtube import YoutubeIE 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 @@ -67,7 +69,7 @@ class VKBaseIE(InfoExtractor): login_page = self._download_webpage( 'https://login.vk.com/?act=login', None, - note='Logging in as %s' % username, + note='Logging in', data=urlencode_postdata(login_form)) if re.search(r'onLoginFailed', login_page): @@ -99,16 +101,16 @@ class VKIE(VKBaseIE): _TESTS = [ { 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521', - 'md5': '0deae91935c54e00003c2a00646315f0', + 'md5': '7babad3b85ea2e91948005b1b8b0cb84', 'info_dict': { 'id': '162222515', - 'ext': 'flv', + '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, }, }, { @@ -117,12 +119,12 @@ class VKIE(VKBaseIE): 'info_dict': { 'id': '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', } }, { @@ -206,10 +208,10 @@ class VKIE(VKBaseIE): 'id': 'V3K4mi0SYkc', 'ext': 'webm', 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate", - 'description': 'md5:d9903938abdc74c738af77f527ca0596', - 'duration': 178, + 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a', + 'duration': 179, 'upload_date': '20130116', - 'uploader': "Children's Joy Foundation", + 'uploader': "Children's Joy Foundation Inc.", 'uploader_id': 'thecjf', 'view_count': int, }, @@ -221,6 +223,7 @@ class VKIE(VKBaseIE): 'id': 'k3lz2cmXyRuJQSjGHUv', 'ext': 'mp4', 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f', + # TODO: fix test by fixing dailymotion description extraction 'description': 'md5:c651358f03c56f1150b555c26d90a0fd', 'uploader': 'AniLibria.Tv', 'upload_date': '20160914', @@ -240,13 +243,16 @@ class VKIE(VKBaseIE): '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, live_mp4 + # finished live stream, postlive_mp4 'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2', 'md5': '90d22d051fccbbe9becfccc615be6791', 'info_dict': { @@ -259,7 +265,7 @@ class VKIE(VKBaseIE): }, }, { - # live stream, hls and rtmp links,most likely already finished live + # live stream, hls and rtmp links, most likely already finished live # stream by the time you are reading this comment 'url': 'https://vk.com/video-140332_456239111', 'only_matching': True, @@ -282,15 +288,24 @@ class VKIE(VKBaseIE): { 'url': 'http://new.vk.com/video205387401_165548505', 'only_matching': True, - } - ] + }, + { + # 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') if video_id: - info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id + info_url = 'https://vk.com/al_video.php?act=show_inline&al=1&video=' + video_id # Some videos (removed?) can only be downloaded with list id specified list_id = mobj.group('list_id') if list_id: @@ -313,9 +328,14 @@ class VKIE(VKBaseIE): '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.' + ERRORS = { r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<': - 'Video %s has been removed from public access due to rightholder complaint.', + 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, ' @@ -329,17 +349,27 @@ class VKIE(VKBaseIE): 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'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) - youtube_url = self._search_regex( - r']+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"', - info_page, 'youtube iframe', default=None) + youtube_url = YoutubeIE._extract_url(info_page) if youtube_url: - return self.url_result(youtube_url, 'Youtube') + return self.url_result(youtube_url, ie=YoutubeIE.ie_key()) vimeo_url = VimeoIE._extract_url(url, info_page) if vimeo_url is not None: @@ -369,27 +399,52 @@ class VKIE(VKBaseIE): opts_url = 'http:' + opts_url return self.url_result(opts_url) - data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars') - data = json.loads(data_json) + # 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*', r'\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] title = unescapeHTML(data['md_title']) - if data.get('live') == 2: + # 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, - '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,.]+)', - info_page, 'view count', fatal=False)) + info_page, 'view count', default=None)) 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'): + 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({ @@ -399,8 +454,8 @@ class VKIE(VKBaseIE): }) elif format_id == 'hls': formats.extend(self._extract_m3u8_formats( - format_url, video_id, 'mp4', m3u8_id=format_id, - fatal=False, live=True)) + format_url, video_id, 'mp4', 'm3u8_native', + m3u8_id=format_id, fatal=False, live=is_live)) elif format_id == 'rtmp': formats.append({ 'format_id': format_id, @@ -415,9 +470,13 @@ class VKIE(VKBaseIE): 'title': title, 'thumbnail': data.get('jpg'), 'uploader': data.get('md_author'), + 'uploader_id': str_or_none(data.get('author_id')), 'duration': data.get('duration'), 'timestamp': timestamp, 'view_count': view_count, + 'like_count': int_or_none(data.get('liked')), + 'dislike_count': int_or_none(data.get('nolikes')), + 'is_live': is_live, }