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'

    remove_empty_feeds =  True
    resolve_internal_links = True
    
    def __init__(self, *args, **kwargs):
        BasicNewsRecipe.__init__(self, *args, **kwargs)
        if self.output_profile.short_name.startswith('kindle'):
            self.title = 'Mint | ' + date.today().strftime('%b %d, %Y')
            if is_saturday:
                self.title = 'Mint Lounge | ' + date.today().strftime('%b %d, %Y')

    if is_saturday:
        
        def get_cover_url(self):
            soup = self.index_to_soup('https://lifestyle.livemint.com/')
            if citem := soup.find('div', attrs={'class':'headLatestIss_cover'}):
                return citem.img['src'].replace('_tn.jpg', '_mr.jpg')
        
        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:
        
        def get_cover_url(self):
            soup = self.index_to_soup(
                'https://www.magzter.com/IN/HT-Digital-Streams-Ltd./Mint-Mumbai/Newspaper/'
            )
            for citem in soup.findAll('meta', content=lambda s: s and s.endswith('view/3.jpg')):
                return citem['content']

        extra_css = '''
            img {display:block; margin:0 auto;}
            #img-cap {font-size:small; text-align:center;}
            .summary, .highlights, .synopsis {
                font-weight:normal !important; font-style:italic; color:#202020;
            }
            h2 {font-size:normal !important;}
            .author-widget {font-size:small; font-style:italic; color:#404040;}
            em, blockquote {color:#202020;}
            .moreAbout, .articleInfo, .metaData, .psTopicsHeading, .topicsTag {font-size:small;}
        '''

        keep_only_tags = [
            dict(name='article', attrs={'id':lambda x: x and x.startswith(('article_', 'box_'))}),
            classes('contentSec')
        ]
        remove_tags = [
            dict(name=['meta', 'link', 'svg', 'button', 'iframe']),
            classes(
                'trendingSimilarHeight moreNews mobAppDownload label msgError msgOk taboolaHeight'
                ' socialHolder imgbig disclamerText disqus-comment-count openinApp2 lastAdSlot bs_logo'
                ' datePublish sepStory premiumSlider moreStory Joinus moreAbout milestone benefitText'
            )
        ]

        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):
            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>'
                raw = re.sub(r'<div class="FirstEle">([^}]*)</div>', body, raw)
                return raw
            else:
                return raw

        def preprocess_html(self, soup):
            for strong in soup.findAll('strong'):
                if strong.find('p'):
                    strong.name = 'div'
            for embed in soup.findAll('div', attrs={'class':'embed'}):
                if nos := embed.find('noscript'):
                    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 span in soup.findAll('span', attrs={'class':'exclusive'}):
                span.extract()
            for img in soup.findAll('img', attrs={'data-src': True}):
                img['src'] = img['data-src']
            if wa := soup.find(**classes('autobacklink-topic')):
                if p := wa.findParent('p'):
                    p.extract()
            return soup

        def populate_article_metadata(self, article, soup, first):
            article.title = article.title.replace('<span class="webrupee">₹</span>','₹')