import re from .common import InfoExtractor from ..utils import ( qualities, unified_timestamp, traverse_obj, ) class PearVideoIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?pearvideo\.com/video_(?P\d+)' _TEST = { 'url': 'http://www.pearvideo.com/video_1076290', 'info_dict': { 'id': '1076290', 'ext': 'mp4', 'title': '小浣熊在主人家玻璃上滚石头:没砸', 'description': 'md5:01d576b747de71be0ee85eb7cac25f9d', 'timestamp': 1494275280, 'upload_date': '20170508', } } def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) quality = qualities( ('ldflv', 'ld', 'sdflv', 'sd', 'hdflv', 'hd', 'src')) formats = [{ 'url': mobj.group('url'), 'format_id': mobj.group('id'), 'quality': quality(mobj.group('id')), } for mobj in re.finditer( r'(?P[a-zA-Z]+)Url\s*=\s*(["\'])(?P(?:https?:)?//.+?)\2', webpage)] if not formats: info = self._download_json( 'https://www.pearvideo.com/videoStatus.jsp', video_id=video_id, query={'contId': video_id}, headers={'Referer': url}) formats = [{ 'format_id': k, 'url': v.replace(info['systemTime'], f'cont-{video_id}') if k == 'srcUrl' else v } for k, v in traverse_obj(info, ('videoInfo', 'videos'), default={}).items() if v] title = self._search_regex( (r']+\bclass=(["\'])video-tt\1[^>]*>(?P[^<]+)', r'<[^>]+\bdata-title=(["\'])(?P(?:(?!\1).)+)\1'), webpage, 'title', group='value') description = self._search_regex( (r']+\bclass=(["\'])summary\1[^>]*>(?P[^<]+)', r'<[^>]+\bdata-summary=(["\'])(?P(?:(?!\1).)+)\1'), webpage, 'description', default=None, group='value') or self._html_search_meta('Description', webpage) timestamp = unified_timestamp(self._search_regex( r']+\bclass=["\']date["\'][^>]*>([^<]+)', webpage, 'timestamp', fatal=False)) return { 'id': video_id, 'title': title, 'description': description, 'timestamp': timestamp, 'formats': formats, }