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 Frontline

Browse Source
This commit is contained in:
Kovid Goyal 2022-07-28 07:22:31 +05:30
parent 7df0187962
commit f0342a3d31
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 8 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
recipes/frontline.recipe
View File

@ -21,17 +21,16 @@ class Frontline(BasicNewsRecipe):
def get_cover_url(self): def get_cover_url(self):
soup = self.index_to_soup( soup = self.index_to_soup(
'https://frontline.thehindu.com/current-issue/') 'https://frontline.thehindu.com/current-issue/')
tag = soup.find(attrs={'class': 'imgWrapper'}) tag = soup.find(attrs={'class': 'sptar-image'})
if tag: if tag:
self.cover_url = tag.find('img')['data-proxy-image'].replace( self.cover_url = tag.find('img')['data-original']
"FREE_100", "FREE_810")
return super().get_cover_url() return super().get_cover_url()
# https://fl.thgim.com/incoming/b5zy2g/article38454943.ece/alternates/FREE_100/coverpng # https://fl.thgim.com/incoming/b5zy2g/article38454943.ece/alternates/FREE_100/coverpng
keep_only_tags = [ keep_only_tags = [
classes( classes(
'overline mainart-title marginBottom10px articleBottomLine swiper-slide slide-caption artlead-text body-main' 'overline mainart-title marginBottom10px articleBottomLine swiper-slide slide-caption artlead-text body-main article-container'
) )
] ]
remove_tags = [classes('dispatche-middle bigtitle')] remove_tags = [classes('dispatche-middle bigtitle')]
@ -60,8 +59,9 @@ class Frontline(BasicNewsRecipe):
] ]
def preprocess_html(self, soup): def preprocess_html(self, soup):
for img in soup.findAll('img', attrs={'data-proxy-image': True}): for source in soup.findAll('source', srcset=True, attrs={'media':'(min-width: 1600px)'}):
# replace FREE_810 with FREE_615 FREE_1200.. FREE_300. source.name = 'img'
img['src'] = img['data-proxy-image'].replace( source['src'] = source['srcset']
"FREE_100", "FREE_810") for img in soup.findAll('img', attrs={'data-original':True}):
img['src'] = img['data-original']
return soup return soup