From 111de00289d8c019764e79247568248f8a4b11f6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?D=C3=A9stin=20Reed?= Date: Sat, 23 Jul 2016 17:02:50 +0200 Subject: [PATCH] [DailyMail] Improve title and description extraction --- youtube_dl/extractor/dailymail.py | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/youtube_dl/extractor/dailymail.py b/youtube_dl/extractor/dailymail.py index b60a1d813..98c835bf1 100644 --- a/youtube_dl/extractor/dailymail.py +++ b/youtube_dl/extractor/dailymail.py @@ -5,19 +5,20 @@ from ..utils import ( int_or_none, determine_protocol, + unescapeHTML, ) class DailyMailIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P[0-9]+)' _TEST = { - 'url': 'http://www.dailymail.co.uk/video/sciencetech/video-1288527/Turn-video-impressionist-masterpiece.html', - 'md5': '2f639d446394f53f3a33658b518b6615', + 'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html', + 'md5': 'f6129624562251f628296c3a9ffde124', 'info_dict': { - 'id': '1288527', + 'id': '1295863', 'ext': 'mp4', - 'title': 'Turn any video into an impressionist masterpiece', - 'description': 'md5:88ddbcb504367987b2708bb38677c9d2', + 'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'', + 'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84', } } @@ -26,7 +27,7 @@ def _real_extract(self, url): webpage = self._download_webpage(url, video_id) video_data = self._parse_json(self._search_regex( r"data-opts='({.+?})'", webpage, 'video data'), video_id) - title = video_data['title'] + title = unescapeHTML(video_data['title']) video_sources = self._download_json(video_data.get( 'sources', {}).get('url') or 'http://www.dailymail.co.uk/api/player/%s/video-sources.json' % video_id, video_id) @@ -55,7 +56,7 @@ def _real_extract(self, url): return { 'id': video_id, 'title': title, - 'description': video_data.get('descr'), + 'description': unescapeHTML(video_data.get('descr')), 'thumbnail': video_data.get('poster') or video_data.get('thumbnail'), 'formats': formats, }