calibre/recipes/hindu.recipe

#!/usr/bin/env python
# vim:fileencoding=utf-8
# License: GPLv3 Copyright: 2009, Kovid Goyal <kovid at kovidgoyal.net>

import string

from calibre import entity_to_unicode
from calibre.web.feeds.news import BasicNewsRecipe


def classes(classes):
    q = frozenset(classes.split(' '))
    return dict(
        attrs={'class': lambda x: x and frozenset(x.split()).intersection(q)})


class TheHindu(BasicNewsRecipe):
    title = u'The Hindu'
    language = 'en_IN'
    epaper_url = 'https://epaper.thehindu.com'

    oldest_article = 1
    __author__ = 'Kovid Goyal'
    max_articles_per_feed = 100
    no_stylesheets = True
    remove_attributes = ['style', 'height', 'width']
    extra_css = '.lead-img-cont { text-align: center; } ' \
                '.lead-img-caption { font-size: small; font-style: italic; } ' \
                '.mobile-author-cont { font-size: small; text-transform: uppercase; } ' \
                '.intro ~ .intro, .update-time, .ksl-time-stamp * { display: none; } '

    ignore_duplicate_articles = {'title', 'url'}
    keep_only_tags = [
        dict(name='h1', attrs={'class': ['title', 'special-article-heading']}),
        classes('lead-img-cont mobile-author-cont photo-collage intro'),
        dict(id=lambda x: x and x.startswith('content-body-')),
    ]

    def get_browser(self):
        br = BasicNewsRecipe.get_browser(self, user_agent='common_words/based')
        br.addheaders += [('Referer', self.epaper_url)]     # needed for fetching cover
        # br.set_debug_http(True)
        return br

    def get_cover_url(self):
        url = self.index_to_soup(self.epaper_url + '/Login/DefaultImage', raw=True)
        return url.replace(br'\\', b'/').decode('utf-8')[1:-1]

    def preprocess_html(self, soup):
        img = soup.find('img', attrs={'class': 'lead-img'})
        try:
            for i, source in enumerate(tuple(img.parent.findAll('source', srcset=True))):
                if i == 0:
                    img['src'] = source['srcset'].split()[0]
                source.extract()
        except Exception:
            pass
        for img in soup.findAll(attrs={'data-original': True}):
            img['src'] = img['data-original']
        # Place intro beneath the title, skip duplicates
        try:
            soup.h1.insert_after(soup.find('h2', attrs={'class': 'intro'}))
        except Exception:
            pass
        # Remove ',' from location tag
        ts = soup.find('span', attrs={'class': 'ksl-time-stamp'})
        if ts and ts.string:
            ts.string = ts.string.split(',')[0]
        return soup

    def populate_article_metadata(self, article, soup, first):
        try:
            desc = soup.find('meta', attrs={'name': 'description'}).get('content')
            if not desc.startswith('Todays paper'):
                desc += '...' if len(desc) >= 199 else ''   # indicate truncation
                article.text_summary = article.summary = entity_to_unicode(desc)
        except AttributeError:
            return

    def articles_from_soup(self, soup):
        ans = []
        div = soup.find('section', attrs={'id': 'section_1'})
        if div is None:
            return ans
        for ul in div.findAll('ul', attrs={'class': 'archive-list'}):
            for x in ul.findAll(['a']):
                title = self.tag_to_string(x)
                url = x.get('href', False)
                if not url or not title:
                    continue
                self.log('\t\tFound article:', title)
                self.log('\t\t\t', url)
                ans.append({
                    'title': title,
                    'url': url,
                    'description': '',
                    'date': ''})
        return ans

    def parse_index(self):
        # return [('xxx', [
        #     {'title':'xxx', 'url':'http://www.thehindu.com/opinion/op-ed/rohingya-bangladeshs-burden-to-bear/article19694058.ece'},
        #     {'title':'yyy', 'url':'http://www.thehindu.com/sci-tech/energy-and-environment/on-river-washed-antique-plains/article19699327.ece'}
        # ])]
        soup = self.index_to_soup('https://www.thehindu.com/todays-paper/')
        nav_div = soup.find(id='subnav-tpbar-latest')
        section_list = []

        # Finding all the section titles that are acceptable
        for x in nav_div.findAll(['a']):
            if self.is_accepted_entry(x):
                section_list.append(
                    (string.capwords(self.tag_to_string(x)), x['href']))
        feeds = []

        # For each section title, fetch the article urls
        for section in section_list:
            section_title = section[0]
            section_url = section[1]
            self.log('Found section:', 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 is_accepted_entry(self, entry):
        # Those sections in the top nav bar that we will omit
        omit_list = [
            'tp-tamilnadu', 'tp-karnataka', 'tp-kerala', 'tp-andhrapradesh',
            'tp-telangana', 'tp-newdelhi', 'tp-mumbai', 'tp-otherstates',
            'tp-in-school', 'tp-metroplus', 'tp-youngworld', 'tp-fridayreview',
            'tp-downtown', 'tp-bookreview', 'tp-others']

        is_accepted = True
        for omit_entry in omit_list:
            if entry['href'][0:-1].endswith(omit_entry):
                is_accepted = False
                break
        return is_accepted