~ubuntu-branches/debian/sid/calibre/sid

#!/usr/bin/env  python

 

__license__   = 'GPL v3'

__copyright__ = u'2010-2013, Tomasz Dlugosz <tomek3d@gmail.com>'

'''

fronda.pl

'''

 

import re

from calibre.web.feeds.news import BasicNewsRecipe

from datetime import timedelta, date

 

class Fronda(BasicNewsRecipe):

    title          = u'Fronda.pl'

    publisher      = u'Fronda.pl'

    description    = u'Portal po\u015bwi\u0119cony - Informacje'

    language = 'pl'

    __author__ = u'Tomasz D\u0142ugosz'

    oldest_article = 7

    max_articles_per_feed = 100

    use_embedded_content = False

    no_stylesheets = True

 

    extra_css = '''

        h1 {font-size:150%}

        .body {text-align:left;}

        div#featured-image {font-style:italic; font-size:70%}

    '''

 

    earliest_date = date.today() - timedelta(days=oldest_article)

 

    def date_cut(self,datestr):

        # eg. 5.11.2012, 12:07

        timestamp = datestr.split(',')[0]

        parts = timestamp.split('.')

        art_date = date(int(parts[2]),int(parts[1]),int(parts[0]))

        return True if art_date < self.earliest_date else False

 

    def parse_index(self):

        genres = [

            ('ekonomia,4.html', 'Ekonomia'),

            ('filozofia,15.html', 'Filozofia'),

            ('historia,6.html', 'Historia'),

            ('kosciol,8.html', 'Kościół'),

            ('kultura,5.html', 'Kultura'),

            ('media,10.html', 'Media'),

            ('nauka,9.html', 'Nauka'),

            ('polityka,11.html', 'Polityka'),

            ('polska,12.html', 'Polska'),

            ('prolife,3.html', 'Prolife'),

            ('religia,7.html', 'Religia'),

            ('rodzina,13.html', 'Rodzina'),

            ('swiat,14.html', 'Świat'),

            ('wydarzenie,16.html', 'Wydarzenie')

        ]

        feeds = []

        articles = {}

 

        for url, genName in genres:

            try:

                soup = self.index_to_soup('http://www.fronda.pl/c/'+ url)

            except:

                continue

            articles[genName] = []

            for item in soup.findAll('li'):

                article_h = item.find('h2')

                if not article_h:

                    continue

                article_date = self.tag_to_string(item.find('b'))

                if self.date_cut(article_date):

                    continue

                article_a = article_h.find('a')

                article_url = 'http://www.fronda.pl' + article_a['href']

                article_title = self.tag_to_string(article_a)

                articles[genName].append( { 'title' : article_title, 'url' : article_url, 'date' : article_date })

            if articles[genName]:

                feeds.append((genName, articles[genName]))

        return feeds

 

    keep_only_tags = [

        dict(name='div', attrs={'class':'yui-g'})

        ]

 

    remove_tags = [

        dict(name='div', attrs={'class':['related-articles','button right','pagination','related-articles content']}),

        dict(name='h3', attrs={'class':'block-header article comments'}),

        dict(name='ul', attrs={'class':['comment-list','category','tag-list']}),

        dict(name='p', attrs={'id':'comments-disclaimer'}),

        dict(name='div', attrs={'style':'text-align: left; margin-bottom: 15px;'}),

        dict(name='div', attrs={'style':'text-align: left; margin-top: 15px; margin-bottom: 30px;'}),

        dict(name='div', attrs={'id':'comment-form'}),

        dict(name='span', attrs={'class':'separator'})

        ]

 

    preprocess_regexps = [

        (re.compile(r'komentarzy: .*?</h6>', re.IGNORECASE | re.DOTALL | re.M ), lambda match: '</h6>')]