calibre/recipes/ieeespectrum.recipe

__license__   = 'GPL v3'
__copyright__ = '2010, Franco Venturi <fventuri at comcast.net>'
'''
spectrum.ieee.org
'''

from calibre.web.feeds.news import BasicNewsRecipe
from string import capwords
from urlparse import urljoin

class IEEESpectrum(BasicNewsRecipe):
    title                 = 'IEEE Spectrum'
    __author__            = 'Franco Venturi'
    description           = 'Electronics News from IEEE'
    publisher             = 'IEEE'
    category              = 'news, electronics, IT, computer science'
    oldest_article        = 32
    max_articles_per_feed = 100
    no_stylesheets        = True
    use_embedded_content  = False
    language              = 'en'
    index                 = 'http://spectrum.ieee.org/magazine/'
    masthead_url          = 'http://spectrum.ieee.org/images/logo_hdr.png'

    remove_javascript     = True
    remove_tags           = [dict(name={'script':True, 'object':True})]
    remove_attributes     = ['height','width','alt']
    keep_only_tags        = [dict(attrs={'class': {'artSctn':True, 'artTitle':True, 'dekTitle': True}}), dict(attrs={'id':'artBody'})]


    def parse_index(self):
        soup = self.index_to_soup(self.index)
        img = soup.find('img', image='cover.gif', src=True)
        if img is not None:
            self.cover_url = 'http://spectrum.ieee.org'+img['src']

        content = soup.find(id='gnrlContent')
        title = content.find(attrs={'class':'style4'}).string.strip()
        date = ' '.join(title.split()[0:2])
        self.timefmt = ' [' + date + ']'
        contents = []
        for tag in content.findAll(attrs={'class': {'style2':True, 'lstngTitle':True, 'lstngBody': True}}):
            if tag['class'] == 'style2':
                contents.append((capwords(tag.renderContents().strip()), []))
            elif tag['class'] == 'lstngTitle':
                url = urljoin(self.index, tag.findPrevious('a')['href']) + '/0'
                contents[-1][1].append({'title': tag.renderContents().strip(),
                                        'url': url,
                                        'date': date,
                                        'description': '',
                                        'content': ''
                                       })
            elif tag['class'] == 'lstngBody':
                contents[-1][1][-1]['description'] = tag.renderContents().strip()

        return contents

    def preprocess_html(self, soup):
        for a in soup.findAll('a'):
            if not a['href'].lower().startswith('http'):
               a['href'] = urljoin(self.index, a['href'])
        return soup