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 livemint.recipe

Browse Source
This commit is contained in:
unkn0w7n 2024-03-05 10:16:40 +05:30
parent 7059a322a0
commit 611370e719
1 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
recipes/livemint.recipe
View File

@ -83,7 +83,7 @@ class LiveMint(BasicNewsRecipe):
return citem['content']
extra_css = '''
img {display:block; margin:0 auto;}
img {margin:0 auto;}
#img-cap {font-size:small; text-align:center;}
.summary, .highlights, .synopsis {
font-weight:normal !important; font-style:italic; color:#202020;
@ -128,6 +128,8 @@ class LiveMint(BasicNewsRecipe):
]
def preprocess_raw_html(self, raw, *a):
# remove empty p tags
raw = re.sub(r'(<p>\s*)(<[^\/])', '\g<2>', re.sub(r'(<p>\s*&nbsp;\s*<\/p>)|(<p>\s*<\/p>)', '', raw))
if '<script>var wsjFlag=true;</script>' in raw:
m = re.search(r'type="application/ld\+json">[^<]+?"@type": "NewsArticle"', raw)
raw1 = raw[m.start():]
@ -137,8 +139,8 @@ class LiveMint(BasicNewsRecipe):
body = data['articleBody'] + '</p> <p>'\
+ re.sub(r'(([a-z]|[^A-Z])\.|\.”)([A-Z]|“[A-Z])', r'\1 <p> \3', value)
body = '<div class="FirstEle"> <p>' + body + '</p> </div>'
raw = re.sub(r'<div class="FirstEle">([^}]*)</div>', body, raw)
return raw
raw2 = re.sub(r'<div class="FirstEle">([^}]*)</div>', body, raw)
return raw2
else:
return raw