Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Update The Australian

Browse Source
This commit is contained in:
Kovid Goyal 2020-12-23 10:39:43 +05:30
parent c9193e3e53
commit de07a03eb7
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 17 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
recipes/the_oz.recipe
View File

@ -33,7 +33,10 @@ class DailyTelegraph(BasicNewsRecipe):
remove_empty_feeds = True remove_empty_feeds = True
ignore_duplicate_articles = {'url'} ignore_duplicate_articles = {'url'}
keep_only_tags = dict(id=['story', 'story-headline']) keep_only_tags = [
classes('story-headline'),
dict(id='story'),
]
remove_tags = [ remove_tags = [
dict(id='comments'), dict(id='comments'),
classes('story-info story-header-tools module-controls story-sidebar' classes('story-info story-header-tools module-controls story-sidebar'
@ -43,14 +46,14 @@ class DailyTelegraph(BasicNewsRecipe):
] ]
feeds = [ feeds = [
(u'National News', u'http://www.news.com.au/national/rss'), (u'National News', u'https://www.news.com.au/content-feeds/latest-news-national/'),
(u'World News', u'http://www.news.com.au/world/rss'), (u'World News', u'https://www.news.com.au/content-feeds/latest-news-world/'),
(u'Entertainmnet', u'http://www.news.com.au/entertainment/rss'), (u'Entertainment', u'https://www.news.com.au/content-feeds/latest-news-entertainment/'),
(u'Technology', u'http://www.news.com.au/technology/rss'), (u'Technology', u'https://www.news.com.au/content-feeds/latest-news-technology/'),
(u'Lifestyle', u'http://www.news.com.au/lifestyle/rss'), (u'Lifestyle', u'https://www.news.com.au/content-feeds/latest-news-lifestyle/'),
(u'Sport', u'http://www.news.com.au/sport/rss'), (u'Sport', u'https://www.news.com.au/content-feeds/latest-news-sport/'),
(u'Finance', u'http://www.news.com.au/finance/rss'), (u'Finance', u'https://www.news.com.au/content-feeds/latest-news-finance/'),
(u'Travel', u'http://www.news.com.au/travel/rss'), (u'Travel', u'https://www.news.com.au/content-feeds/latest-news-travel/'),
] ]
def get_article_url(self, article): def get_article_url(self, article):
@ -58,3 +61,8 @@ class DailyTelegraph(BasicNewsRecipe):
if '/video/' in ans: if '/video/' in ans:
return return
return ans return ans
def preprocess_html(self, soup):
for img in soup.findAll('img', attrs={'data-src': True}):
img['src'] = img['data-src']
return soup