Wisconsin Public Radio Calibre News Feed Recipe

wpr.recipe

#!/usr/bin/env python
# vim:fileencoding=utf-8
from calibre.web.feeds.recipes import BasicNewsRecipe

class WPR(BasicNewsRecipe):
    title          = 'Wisconsin Public Radio'
    author         = 'V David Zvenyach'
    description    = 'Daily news from the Wisconsin Public Radio'
    no_stylesheets = True
    compress_news_images = True
    keep_only_tags = [dict(name='div', attrs={'class':'block-main'})]
    remove_tags= [
        dict(name='div', attrs={'class':'block-inject'}),
        dict(name='div', attrs={'class':'field-name-service-links-displays-group'})
    ]

    def parse_index(self):
        soup = self.index_to_soup('https://www.wpr.org/news/index.html')
        articles = [
        ]
        for div in soup.findAll('div', attrs={'class':'views-row'}):
            h2 = div.find('h2')
            if not h2:
                     continue
            url = "https://www.wpr.org" + h2.a.get('href')
            title = h2.get_text()
            date = div.find('span',attrs={'class':'date-display-single'}).get_text()
            description = ""
            content = ""
            articles.append(dict(
                url=url,
                title=title,
                date=date,
                description=description, 
                content=content
            ))
        return [('WPR News', articles)]