Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-07 10:14:46 -04:00
Code Issues Packages Projects Releases Wiki Activity

Update Washington Post

Browse Source
This commit is contained in:
Kovid Goyal 2020-05-06 07:46:43 +05:30
parent 876623ab2e
commit 450b64a6f8
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
recipes/wash_post.recipe
View File

@ -29,11 +29,13 @@ class TheWashingtonPost(BasicNewsRecipe):
publication_type = 'newspaper' publication_type = 'newspaper'
keep_only_tags = [ keep_only_tags = [
dict(itemprop=['headline', 'articleBody']), dict(name=['h1', 'figure']),
classes('byline article-body'),
] ]
remove_tags = [ remove_tags = [
dict(name=['meta', 'link']), dict(name=['meta', 'link']),
classes('inline-video'), classes('inline-video author-tooltip author-image'),
dict(attrs={'data-qa': 'article-body-ad'}),
] ]
feeds = [ feeds = [
@ -51,6 +53,6 @@ class TheWashingtonPost(BasicNewsRecipe):
] ]
def preprocess_html(self, soup, *a): def preprocess_html(self, soup, *a):
for img in soup.findAll('img', attrs={'data-low-res-src': True}): for img in soup.findAll('img', srcset=True):
img['src'] = img['data-low-res-src'] img['src'] = img['srcset'].split()[0]
return soup return soup