import re from .common import InfoExtractor from .youtube import YoutubeIE from ..utils import ( parse_iso8601, str_to_int, ) class CrackedIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?cracked\.com/video_(?P\d+)_[\da-z-]+\.html' _TESTS = [{ 'url': 'http://www.cracked.com/video_19070_if-animal-actors-got-e21-true-hollywood-stories.html', 'md5': '89b90b9824e3806ca95072c4d78f13f7', 'info_dict': { 'id': '19070', 'ext': 'mp4', 'title': 'If Animal Actors Got E! True Hollywood Stories', 'timestamp': 1404954000, 'upload_date': '20140710', }, }, { # youtube embed 'url': 'http://www.cracked.com/video_19006_4-plot-holes-you-didnt-notice-in-your-favorite-movies.html', 'md5': 'ccd52866b50bde63a6ef3b35016ba8c7', 'info_dict': { 'id': 'EjI00A3rZD0', 'ext': 'mp4', 'title': "4 Plot Holes You Didn't Notice in Your Favorite Movies - The Spit Take", 'description': 'md5:c603708c718b796fe6079e2b3351ffc7', 'upload_date': '20140725', 'uploader_id': 'Cracked', 'uploader': 'Cracked', }, }] def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) youtube_url = YoutubeIE._extract_url(webpage) if youtube_url: return self.url_result(youtube_url, ie=YoutubeIE.ie_key()) video_url = self._html_search_regex( [r'var\s+CK_vidSrc\s*=\s*"([^"]+)"', r'([^<]+)'], webpage, 'title') description = self._search_regex( r'name="?(?:og:)?description"?\s+content="([^"]+)"', webpage, 'description', default=None) timestamp = self._html_search_regex( r'"date"\s*:\s*"([^"]+)"', webpage, 'upload date', fatal=False) if timestamp: timestamp = parse_iso8601(timestamp[:-6]) view_count = str_to_int(self._html_search_regex( r'([\d,\.]+) Views', webpage, 'view count', fatal=False)) comment_count = str_to_int(self._html_search_regex( r'([\d,\.]+)', webpage, 'comment count', fatal=False)) m = re.search(r'_(?P\d+)X(?P\d+)\.mp4$', video_url) if m: width = int(m.group('width')) height = int(m.group('height')) else: width = height = None return { 'id': video_id, 'url': video_url, 'title': title, 'description': description, 'timestamp': timestamp, 'view_count': view_count, 'comment_count': comment_count, 'height': height, 'width': width, }