calibre/recipes/hbr.recipe

from calibre.web.feeds.news import BasicNewsRecipe
import re
from datetime import date, timedelta

class HBR(BasicNewsRecipe):

    title = 'Harvard Business Review'
    description = 'To subscribe go to http://hbr.harvardbusiness.org'
    needs_subscription = True
    __author__ = 'Kovid Goyal and Sujata Raman'
    timefmt                = ' [%B %Y]'
    language = 'en'
    no_stylesheets = True
    recipe_disabled = ('hbr.org has started requiring the use of javascript'
            ' to log into their website. This is unsupported in calibre, so'
            ' this recipe has been disabled. If you would like to see '
            ' HBR supported in calibre, contact hbr.org and ask them'
            ' to provide a javascript free login method.')

    LOGIN_URL = 'https://hbr.org/login?request_url=/'
    LOGOUT_URL = 'https://hbr.org/logout?request_url=/'

    INDEX = 'http://hbr.org/archive-toc/BR'

    keep_only_tags = [dict(name='div', id='pageContainer')]
    remove_tags = [dict(id=['mastheadContainer', 'magazineHeadline',
        'articleToolbarTopRD', 'pageRightSubColumn', 'pageRightColumn',
        'todayOnHBRListWidget', 'mostWidget', 'keepUpWithHBR',
        'mailingListTout', 'partnerCenter', 'pageFooter',
        'superNavHeadContainer', 'hbrDisqus',
        'articleToolbarTop', 'articleToolbarBottom', 'articleToolbarRD']),
        dict(name='iframe')]
    extra_css = '''
                a {font-family:Georgia,"Times New Roman",Times,serif; font-style:italic; color:#000000; }
                .article{font-family:Georgia,"Times New Roman",Times,serif; font-size: xx-small;}
                h2{font-family:Georgia,"Times New Roman",Times,serif; font-weight:bold; font-size:large; }
                h4{font-family:Georgia,"Times New Roman",Times,serif; font-weight:bold; font-size:small;  }
                #articleAuthors{font-family:Georgia,"Times New Roman",Times,serif; font-style:italic; color:#000000;font-size:x-small;}
                #summaryText{font-family:Georgia,"Times New Roman",Times,serif; font-weight:bold; font-size:x-small;}
                '''

    def get_browser(self):
        br = BasicNewsRecipe.get_browser(self)
        self.logout_url = None

        #'''
        br.open(self.LOGIN_URL)
        br.select_form(name='signin-form')
        br['signin-form:username'] = self.username
        br['signin-form:password'] = self.password
        raw = br.submit().read()
        if '>Sign out<' not in raw:
            raise Exception('Failed to login, are you sure your username and password are correct?')
        try:
            link = br.find_link(text='Sign out')
            if link:
                self.logout_url = link.absolute_url
        except:
            self.logout_url = self.LOGOUT_URL
        #'''
        return br

    def cleanup(self):
        if self.logout_url is not None:
            self.browser.open(self.logout_url)

    def map_url(self, url):
        if url.endswith('/ar/1'):
            return url[:-1]+'pr'


    def hbr_get_toc(self):
        #return self.index_to_soup(open('/t/hbr.html').read())

        today = date.today()
        future = today + timedelta(days=30)
        for x in [x.strftime('%y%m') for x in (future, today)]:
            url = self.INDEX + x
            soup = self.index_to_soup(url)
            if not soup.find(text='Issue Not Found'):
                return soup
        raise Exception('Could not find current issue')

    def hbr_parse_toc(self, soup):
        feeds = []
        current_section = None
        articles = []
        for x in soup.find(id='archiveToc').findAll(['h3', 'h4']):
            if x.name == 'h3':
                if current_section is not None and articles:
                    feeds.append((current_section, articles))
                current_section = self.tag_to_string(x).capitalize()
                articles = []
                self.log('\tFound section:', current_section)
            else:
                a = x.find('a', href=True)
                if a is None: continue
                title = self.tag_to_string(a)
                url = a['href']
                if '/ar/' not in url:
                    continue
                if url.startswith('/'):
                    url = 'http://hbr.org' + url
                url = self.map_url(url)
                p = x.parent.find('p')
                desc = ''
                if p is not None:
                    desc = self.tag_to_string(p)
                self.log('\t\tFound article:', title)
                self.log('\t\t\t', url)
                self.log('\t\t\t', desc)

                articles.append({'title':title, 'url':url, 'description':desc,
                    'date':''})
        return feeds


    def parse_index(self):
        soup = self.hbr_get_toc()
        #open('/t/hbr.html', 'wb').write(unicode(soup).encode('utf-8'))
        feeds = self.hbr_parse_toc(soup)
        return feeds

    def get_cover_url(self):
        cover_url = None
        index = 'http://hbr.org/current'
        soup = self.index_to_soup(index)
        link_item = soup.find('img', alt=re.compile("Current Issue"), src=True)

        if link_item:
           cover_url = 'http://hbr.org' + link_item['src']

        return cover_url