[DailyMail] Improve title and description extraction

This commit is contained in:
Déstin Reed 2016-07-23 17:02:50 +02:00 committed by Sergey M․
parent b4a131e1a5
commit 111de00289
No known key found for this signature in database
GPG key ID: 2C393E0F18A9236D

View file

@ -5,19 +5,20 @@
from ..utils import ( from ..utils import (
int_or_none, int_or_none,
determine_protocol, determine_protocol,
unescapeHTML,
) )
class DailyMailIE(InfoExtractor): class DailyMailIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P<id>[0-9]+)' _VALID_URL = r'https?://(?:www\.)?dailymail\.co\.uk/video/[^/]+/video-(?P<id>[0-9]+)'
_TEST = { _TEST = {
'url': 'http://www.dailymail.co.uk/video/sciencetech/video-1288527/Turn-video-impressionist-masterpiece.html', 'url': 'http://www.dailymail.co.uk/video/tvshowbiz/video-1295863/The-Mountain-appears-sparkling-water-ad-Heavy-Bubbles.html',
'md5': '2f639d446394f53f3a33658b518b6615', 'md5': 'f6129624562251f628296c3a9ffde124',
'info_dict': { 'info_dict': {
'id': '1288527', 'id': '1295863',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Turn any video into an impressionist masterpiece', 'title': 'The Mountain appears in sparkling water ad for \'Heavy Bubbles\'',
'description': 'md5:88ddbcb504367987b2708bb38677c9d2', 'description': 'md5:a93d74b6da172dd5dc4d973e0b766a84',
} }
} }
@ -26,7 +27,7 @@ def _real_extract(self, url):
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
video_data = self._parse_json(self._search_regex( video_data = self._parse_json(self._search_regex(
r"data-opts='({.+?})'", webpage, 'video data'), video_id) 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( 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) '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 { return {
'id': video_id, 'id': video_id,
'title': title, 'title': title,
'description': video_data.get('descr'), 'description': unescapeHTML(video_data.get('descr')),
'thumbnail': video_data.get('poster') or video_data.get('thumbnail'), 'thumbnail': video_data.get('poster') or video_data.get('thumbnail'),
'formats': formats, 'formats': formats,
} }