#!/usr/bin/env python # vim:fileencoding=utf-8 from __future__ import unicode_literals, division, absolute_import, print_function from calibre.web.feeds.news import BasicNewsRecipe # author: class LePeupleBreton(BasicNewsRecipe): title = 'Le Peuple Breton' __author__ = 'Lionel Plais' description = u'Aujourd\'hui, être libre c\'est être informé' oldest_article = 90 language = 'fr' cover_img_url = 'http://lepeuplebreton.bzh/wp-content/uploads/2017/11/le-peuple-breton-logo.jpg' masthead_url = cover_img_url remove_javascript = True no_stylesheets = True keep_only_tags = [ dict(name='article'), ] remove_tags = [ dict(name='div', attrs={'class': ['share-post', 'juiz_sps_links', 'counters_both', 'juiz_sps_displayed_bottom']}), ] feeds = [ (u'À la Une', u'http://lepeuplebreton.bzh/category/a-la-une/feed/'), (u'Bretagne', u'http://lepeuplebreton.bzh/category/themes/bretagne/feed/'), (u'France', u'http://lepeuplebreton.bzh/category/lieu/france/feed/'), (u'International', u'http://lepeuplebreton.bzh/category/lieu/international-fr/feed/'), (u'Point de vue', u'http://lepeuplebreton.bzh/category/themes/point-de-vue/feed/'), (u'Économie', u'https://lepeuplebreton.bzh/category/themes/economie/feed/'), (u'Écologie', u'http://lepeuplebreton.bzh/category/themes/ecologie/feed/'), (u'Santé', u'http://lepeuplebreton.bzh/category/themes/sante/feed/'), (u'Culture & Sport', u'http://lepeuplebreton.bzh/category/themes/culture/feed/'), ] def preprocess_html(self, soup): for alink in soup.findAll('a'): if alink.string is not None: tstr = alink.string alink.replaceWith(tstr) return soup