calibre/recipes/ap.recipe

from calibre.web.feeds.news import BasicNewsRecipe


class AssociatedPress(BasicNewsRecipe):

    title = u'Associated Press'
    description = 'Global news'
    __author__ = 'Krittika Goyal'
    use_embedded_content = False
    language = 'en'
    no_stylesheets = True
    conversion_options = {
        'linearize_tables': True
    }
    keep_only_tags = {'name': 'table', 'attrs': {
        'class': lambda x: x and 'ap-story-table' in x.split()}}
    remove_tags = [
        {'class': ['ap-mediabox-table']},
        {'name': 'img', 'src': lambda x: x and '//analytics.' in x},
    ]

    def parse_index(self):
        feeds = []
        fronts = ('HOME', 'US', 'WORLD', 'BUSINESS', 'TECHNOLOGY',
                  'SPORTS', 'ENTERTAINMENT', 'HEALTH', 'SCIENCE', 'POLITICS')
        for front in fronts:
            feeds.append([front.capitalize(), self.parse_section(front)])
        feeds[0][0] = 'Top Stories'
        return feeds

    def parse_section(self, front):
        self.log('Processing section:', front)
        soup = self.index_to_soup(
            'http://hosted.ap.org/dynamic/fronts/%s?SITE=AP' % front)

        articles = []
        for x in soup.findAll('p', attrs={'class': ['ap-newsbriefitem-p', 'ap-topheadlineitem-p']}):
            a = x.find('a', href=True)
            title = self.tag_to_string(a)
            url = "http://hosted.ap.org" + a['href']
            p = x.find(attrs={'class': 'topheadlinebody'})
            desc = ''
            if p is not None:
                desc = self.tag_to_string(p)
            self.log('\tFound article:', title, '\n\t\t', desc)
            articles.append({'title': title, 'url': url})

        self.log('\n\n')

        return articles