import re
from calibre.web.feeds.recipes import BasicNewsRecipe

class JournalofHospitalMedicine(BasicNewsRecipe):

    title       = 'Journal of Hospital Medicine'
    __author__  = 'Kovid Goyal'
    description = 'Medical news'
    timefmt = ' [%d %b, %Y]'
    needs_subscription = True
    language = 'en'

    no_stylesheets = True
    keep_only_tags = [dict(id=['articleTitle', 'articleMeta', 'fulltext'])]
    remove_tags = [dict(attrs={'class':'licensedContent'})]


   # TO LOGIN
    def get_browser(self):
        br = BasicNewsRecipe.get_browser()
        br.open('http://www3.interscience.wiley.com/cgi-bin/home')
        br.select_form(nr=0)
        br['j_username'] = self.username
        br['j_password'] = self.password
        response = br.submit()
        raw = response.read()
        if '<h2>LOGGED IN</h2>' not in raw:
            raise Exception('Login failed. Check your username and password')
        return br

    #TO GET ARTICLE TOC
    def johm_get_index(self):
        return self.index_to_soup('http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1553-5606/currentissue')

    # To parse artice toc
    def parse_index(self):
        soup = self.johm_get_index()
        toc = soup.find(id='issueTocGroups')
        feeds = []
        for group in toc.findAll('li', id=re.compile(r'group\d+')):
            gtitle = group.find(attrs={'class':'subSectionHeading'})
            if gtitle is None:
                continue
            gtitle = self.tag_to_string(gtitle)
            arts = group.find(attrs={'class':'articles'})
            if arts is None:
                continue
            self.log('Found section:', gtitle)
            articles = []
            for art in arts.findAll(attrs={'class':lambda x: x and 'tocArticle'
                in x}):
                a = art.find('a', href=True)
                if a is None:
                    continue
                url = a.get('href')
                if url.startswith('/'):
                    url = 'http://onlinelibrary.wiley.com' + url
                url = url.replace('/abstract', '/full')
                title = self.tag_to_string(a)
                a.extract()
                pm = art.find(attrs={'class':'productMenu'})
                if pm is not None:
                    pm.extract()
                desc = self.tag_to_string(art)
                self.log('\tFound article:', title, 'at', url)
                articles.append({'title':title, 'url':url, 'description':desc,
                    'date':''})
            if articles:
                feeds.append((gtitle, articles))

        return feeds