calibre/recipes/eenadu.recipe

from calibre.web.feeds.news import BasicNewsRecipe, classes
from datetime import date, datetime, timedelta
from calibre.utils.date import parse_date
import re


class eenadu_ts(BasicNewsRecipe):
    title = 'ఈనాడు - తెలంగాణ'
    __author__ = 'unkn0wn'
    description = 'THE LARGEST CIRCULATED TELUGU DAILY'
    language = 'te'
    use_embedded_content = False
    remove_javascript = True
    no_stylesheets = True
    remove_attributes = ['height', 'width', 'style']
    ignore_duplicate_articles = {'url', 'title'}
    masthead_url = 'https://dxxd96tbpm203.cloudfront.net//img/logo.png'
    cover_url = 'https://d66zsp32hue2v.cloudfront.net/Eenadu/2022/08/08/GTH/5_01/d5041804_01_mr.jpg'
    encoding = 'utf-8'
    remove_empty_feeds = True
    extra_css = '.pub-t{font-size:small; font-style:italic;}'

    keep_only_tags = [
        dict(name='h1'),
        dict(**classes('pub-t')),
        classes('fullstory text-justify contlist-cont'),
        dict(name='span', attrs={'id': 'PDSAIApbreak'}),
    ]

    remove_tags = [
        dict(name='span', attrs={'style': 'float:left; margin-right:10px;'}),
        dict(
            name='p',
            attrs={
                'style':
                'font-size: 18px !important; margin: 0px; margin-top: -15px; text-align: center;flex: 1;'
            }
        ),
        dict(name='aside', attrs={'class': lambda x: x and x.startswith('thumb')}),
        dict(name='br'),
        classes('sshare-c tags andbeyond_ad fnt20 arti more2 offset-tb1 msb-list')
    ]

    def parse_index(self):
        section_list = [
            ('తెలంగాణ తాజా వార్తలు', 'telangana'),
            ('సంపాదకీయం', 'telangana/editorial'),
            ('తెలంగాణ ప్రధానాంశాలు', 'telangana/top-news'),
            ('తెలంగాణ జిల్లా వార్తలు', 'telangana/districts'),
            # ('క్రైమ్', 'crime'),
            ('పాలిటిక్స్', 'politics'),
            ('జాతీయం', 'india'),
            ('బిజినెస్', 'business'),
            ('అంతర్జాతీయం', 'world'),
            ('క్రీడలు', 'sports'),
            # ('సినిమా', 'movies'),
            # ('చదువు', 'education'),
            # ('సుఖీభవ', 'health'),
            # ('ఈ-నాడు', 'technology'),
            # ('మకరందం', 'devotional'),
            # ('ఈ తరం', 'youth'),
            # ('ఆహా', 'recipes'),
            # ('హాయ్ బుజ్జీ', 'kids-stories'),
            # ('స్థిరాస్తి', 'real-estate'),
        ]
        is_sunday = date.today().weekday() == 6
        if is_sunday:
            section_list.append(('సండే మ్యాగజైన్', 'sunday-magazine'))
        feeds = []

        # For each section title, fetch the article urls
        for section in section_list:
            section_title = section[0]
            section_url = 'https://www.eenadu.net/' + section[1]
            self.log(section_title, section_url)
            soup = self.index_to_soup(section_url)
            articles = self.articles_from_soup(soup)
            if articles:
                feeds.append((section_title, articles))
        return feeds

    def articles_from_soup(self, soup):
        ans = []
        for link in soup.findAll(
            attrs={
                'class': ['telugu_uni_body', 'thumb-content-more', 'owl-carousel']
            }
        ):
            for a in link.findAll('a', attrs={'href': True}):
                url = a['href']
                if url.startswith('https') is False:
                    url = 'https://www.eenadu.net/' + url

                try:
                    desc = self.tag_to_string(a.find('div')).strip()
                except Exception:
                    desc = ''

                for h3 in a.findAll('h3'):
                    title = self.tag_to_string(h3).strip()
                    sub = re.escape(title)
                    desc = re.sub(sub, '', desc).strip()

                    if not title or not url:
                        continue

                    self.log('\t', title, '\n\t', desc, '\n\t\t', url)
                    ans.append({'title': title, 'url': url, 'description': desc})
        return ans

    def preprocess_html(self, soup):
        div = soup.find('div', **classes('pub-t'))
        if div:
            date = parse_date(
                self.tag_to_string(div).strip().replace('Published : ', '').replace(
                    'Updated : ', ''
                ).replace(' IST', ':00.000001')
            ).replace(tzinfo=None)
            today = datetime.now()
            if (today - date) > timedelta(1.15):
                self.abort_article('Skipping old article')
        else:
            self.abort_article('not an article')
        for img in soup.findAll('img', attrs={'data-src': True}):
            img['src'] = img['data-src']
        return soup