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 theeconomictimes_india_print_edition.recipe

Browse Source
This commit is contained in:
unkn0w7n 2023-08-26 13:15:29 +05:30
parent a9b4c3f4db
commit 37e21f78ee
1 changed files with 8 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
recipes/theeconomictimes_india_print_edition.recipe
View File

@ -22,8 +22,9 @@ class TheEconomicTimes(BasicNewsRecipe):
masthead_url = 'https://upload.wikimedia.org/wikipedia/commons/9/98/The_Economic_Times_logo.svg' masthead_url = 'https://upload.wikimedia.org/wikipedia/commons/9/98/The_Economic_Times_logo.svg'
ignore_duplicate_articles = {'title', 'url'} ignore_duplicate_articles = {'title', 'url'}
extra_css = ''' extra_css = '''
.summary {color:#404040; font-style:italic;} .artByline {font-size:small;}
time{font-size:small;} .artImg, .imgBox {font-size:small; color:#202020; text-align:center;}
img {display:block; margin:0 auto;}
''' '''
def get_cover_url(self): def get_cover_url(self):
@ -45,13 +46,14 @@ class TheEconomicTimes(BasicNewsRecipe):
keep_only_tags = [ keep_only_tags = [
dict(name='h1'), dict(name='h1'),
classes( classes(
'artByline artSyn artImg artText publisher publish_on slideshowPackage' 'artByline pageContent'
), ),
] ]
remove_tags = [ remove_tags = [
dict(name='button'),
classes( classes(
'story_title storyCollection shareBar sr_widget_free jsSrWidgetFree srwidgetfree_3' 'story_title storyCollection shareBar sr_widget_free jsSrWidgetFree srwidgetfree_3 showWhatsMsg artSyn'
' sr_paid jsSrWidgetPaid ar_wrp arwd_ld_chk adBox custom_ad mgid orn_free_r bold' ' sr_paid jsSrWidgetPaid ar_wrp arwd_ld_chk adBox custom_ad mgid orn_free_r bold primeSWrapper external_widget'
), ),
] ]
@ -95,11 +97,8 @@ class TheEconomicTimes(BasicNewsRecipe):
return feeds return feeds
def preprocess_html(self, soup): def preprocess_html(self, soup):
h2 = soup.find(**classes('summary'))
if h2:
h2.name = 'p'
for image in soup.findAll('img', attrs={'src': True}): for image in soup.findAll('img', attrs={'src': True}):
image['src'] = image['src'].replace("width-300", "width-640") image['src'] = image['src'].replace("width-300", "width-640")
for img in soup.findAll('img', attrs={'data-original': True}): for img in soup.findAll('img', attrs={'data-original': True}):
img['src'] = img['data-original'].replace('photo', 'thumb').replace('quality-100', 'quality-100,width-600,resizemode-4') img['src'] = img['data-original']
return soup return soup