calibre/recipes/livemint.recipe
unkn0w7n 76684b3a2b ...
2024-07-25 12:49:25 +05:30

199 lines
8.7 KiB
Python

#!/usr/bin/env python
# vim:fileencoding=utf-8
import json
import re
from datetime import date
from calibre.web.feeds.news import BasicNewsRecipe, classes
is_saturday = date.today().weekday() == 5
class LiveMint(BasicNewsRecipe):
title = 'Live Mint'
description = 'Financial News from India.'
language = 'en_IN'
__author__ = 'Krittika Goyal, revised by unkn0wn'
oldest_article = 1.15 # days
max_articles_per_feed = 50
encoding = 'utf-8'
use_embedded_content = False
no_stylesheets = True
remove_attributes = ['style', 'height', 'width']
masthead_url = 'https://images.livemint.com/static/livemint-logo-v1.svg'
recipe_specific_options = {
'days': {
'short': 'Oldest article to download from this news source. In days ',
'long': 'For example, 0.5, gives you articles from the past 12 hours',
'default': str(oldest_article)
}
}
remove_empty_feeds = True
resolve_internal_links = True
def __init__(self, *args, **kwargs):
BasicNewsRecipe.__init__(self, *args, **kwargs)
d = self.recipe_specific_options.get('days')
if d and isinstance(d, str):
self.oldest_article = float(d)
def get_cover_url(self):
today = date.today().strftime('%d/%m/%Y')
today = today.replace('/', '%2F')
raw = self.index_to_soup(
'https://epaper.livemint.com/Home/GetAllpages?editionid=1&editiondate=' + today, raw=True
)
for cov in json.loads(raw):
if cov['NewsProPageTitle'].lower().startswith(('front', 'cover')):
return cov['HighResolution']
if is_saturday:
title = 'Mint Lounge'
masthead_url = 'https://lifestyle.livemint.com/mintlounge/static-images/lounge-logo.svg'
oldest_article = 6.5 # days
extra_css = '''
#story-summary-0 {font-style:italic; color:#202020;}
.innerBanner, .storyImgSec {text-align:center; font-size:small;}
.author {font-size:small;}
'''
keep_only_tags = [
classes('storyPageHeading storyContent innerBanner author')
]
remove_tags = [
dict(name=['meta', 'link', 'svg', 'button', 'iframe']),
classes('hidden-article-url sidebarAdv similarStoriesClass moreFromSecClass linkStories publishDetail'),
dict(attrs={'id':['hidden-article-id-0', 'hidden-article-type-0']})
]
feeds = [
('Lounge News', 'https://lifestyle.livemint.com/rss/news'),
('Food', 'https://lifestyle.livemint.com/rss/food'),
('Fashion', 'https://lifestyle.livemint.com/rss/fashion'),
('How to Lounge', 'https://lifestyle.livemint.com/rss/how-to-lounge'),
('Smart Living', 'https://lifestyle.livemint.com/rss/smart-living'),
('Health', 'https://lifestyle.livemint.com/rss/health'),
('Relationships', 'https://lifestyle.livemint.com//rss/relationships')
]
def preprocess_html(self, soup):
if h2 := soup.find('h2'):
h2.name = 'p'
for also in soup.findAll('h2'):
if self.tag_to_string(also).strip().startswith('Also'):
also.extract()
for img in soup.findAll('img', attrs={'data-img': True}):
img['src'] = img['data-img']
return soup
else:
extra_css = '''
img {margin:0 auto;}
.psTopLogoItem img, .ecologoStory { width:100; }
#img-cap {font-size:small; text-align:center;}
.summary, .highlights, .synopsis {
font-weight:normal !important; font-style:italic; color:#202020;
}
em, blockquote {color:#202020;}
.moreAbout, .articleInfo, .metaData, .psTopicsHeading, .topicsTag, .auth {font-size:small;}
'''
keep_only_tags = [
dict(name='article', attrs={'id':lambda x: x and x.startswith(('article_', 'box_'))}),
dict(attrs={'class':lambda x: x and x.startswith('storyPage_storyBox__')}),
classes('contentSec')
]
remove_tags = [
dict(name=['meta', 'link', 'svg', 'button', 'iframe']),
dict(attrs={'class':lambda x: x and x.startswith(
('storyPage_alsoRead__', 'storyPage_firstPublishDate__', 'storyPage_bcrumb__')
)}),
dict(attrs={'id':['faqSection', 'seoText', 'ellipsisId']}),
classes(
'trendingSimilarHeight moreNews mobAppDownload label msgError msgOk taboolaHeight gadgetSlider ninSec'
' socialHolder imgbig disclamerText disqus-comment-count openinApp2 lastAdSlot bs_logo author-widget'
' datePublish sepStory premiumSlider moreStory Joinus moreAbout milestone benefitText checkCibilBtn trade'
)
]
feeds = [
('Companies', 'https://www.livemint.com/rss/companies'),
('Opinion', 'https://www.livemint.com/rss/opinion'),
('Money', 'https://www.livemint.com/rss/money'),
('Economy', 'https://www.livemint.com/rss/economy'),
('Politics', 'https://www.livemint.com/rss/politics'),
('Science', 'https://www.livemint.com/rss/science'),
('Industry', 'https://www.livemint.com/rss/industry'),
('Education', 'https://www.livemint.com/rss/education'),
('Sports', 'https://www.livemint.com/rss/sports'),
('Technology', 'https://www.livemint.com/rss/technology'),
('News', 'https://www.livemint.com/rss/news'),
('Mutual Funds', 'https://www.livemint.com/rss/Mutual Funds'),
('Markets', 'https://www.livemint.com/rss/markets'),
('AI', 'https://www.livemint.com/rss/AI'),
('Insurance', 'https://www.livemint.com/rss/insurance'),
('Budget', 'https://www.livemint.com/rss/budget'),
('Elections', 'https://www.livemint.com/rss/elections'),
]
def preprocess_raw_html(self, raw, *a):
# remove empty p tags
raw = re.sub(
r'(<p>\s*)(<[^(\/|a|i|b|em|strong)])', '\g<2>', re.sub(
r'(<p>\s*&nbsp;\s*<\/p>)|(<p>\s*<\/p>)|(<p\s*\S+>&nbsp;\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():]
raw1 = raw1.split('>', 1)[1].strip()
data = json.JSONDecoder().raw_decode(raw1)[0]
value = data['hasPart']['value']
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>'
raw2 = re.sub(r'<div class="FirstEle">([^}]*)</div>', body, raw)
return raw2
return raw
def preprocess_html(self, soup):
for h2 in soup.findAll('h2'):
h2.name = 'h4'
auth = soup.find(attrs={'class':lambda x: x and x.startswith(('storyPage_authorInfo__', 'storyPage_authorSocial__'))})
if auth:
auth['class'] = 'auth'
summ = soup.find(attrs={'class':lambda x: x and x.startswith('storyPage_summary__')})
if summ:
summ['class'] = 'summary'
for strong in soup.findAll('strong'):
if strong.find('p'):
strong.name = 'div'
for embed in soup.findAll('div', attrs={'class':'embed'}):
nos = embed.find('noscript')
if nos:
nos.name = 'span'
for span in soup.findAll('figcaption'):
span['id'] = 'img-cap'
for auth in soup.findAll('span', attrs={'class':lambda x: x and 'articleInfo' in x.split()}):
auth.name = 'div'
for img in soup.findAll('img', attrs={'data-src': True}):
img['src'] = img['data-src']
for span in soup.findAll('span', attrs={'class':'exclusive'}):
span.extract()
for al in soup.findAll('a', attrs={'class':'manualbacklink'}):
pa = al.findParent(['p', 'h2', 'h3', 'h4'])
if pa:
pa.extract()
wa = soup.find(**classes('autobacklink-topic'))
if wa:
p = wa.findParent('p')
if p:
p.extract()
return soup
def populate_article_metadata(self, article, soup, first):
article.title = article.title.replace('<span class="webrupee">₹</span>','')