from calibre.web.feeds.news import BasicNewsRecipe
from css_selectors import Select

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

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

    keep_only_tags = [
        dict(attrs={'class':['article-hed', 'byline']}),
        dict(attrs={'class':lambda x: x and 'article' in x.split()}),
    ]
    remove_tags = [
        dict(name='personalization-placement'),
    ]
    use_javascript_to_login = True

    def javascript_login(self, br, username, password):
        br.visit('https://hbr.org/sign-in')
        br.run_for_a_time(15)
        f = br.select_form('sign-in form')
        f['login-email'] = username
        f['login-password'] = password
        br.submit('[js-target="submit-sign-in"]', wait_for_load=False)
        br.run_for_a_time(15)

    def hbr_parse_toc(self, url):
        root = self.index_to_soup(url, as_tree=True)
        select = Select(root)
        section = 'Unknown'
        articles = []
        feeds = []
        toc = next(select('stream-content[data-stream-name="table-of-contents"] ul'))
        for x in toc.xpath('descendant::*[local-name()="h4" or local-name()="stream-item"]'):
            if 'h4' in x.tag:
                if articles:
                    feeds.append((section, articles))
                section = self.tag_to_string(x)
                articles = []
                self.log('Found section:', section)
            else:
                title, url = x.get('data-title'), x.get('data-url')
                desc = ''.join(c.tail or '' for c in x).strip()
                authors = x.get('data-authors')
                if authors:
                    desc = 'by ' + authors + ': ' + desc
                self.log('\tFound article:', title, url, desc)
                articles.append({'title':title, 'url':'https://hbr.org' + url, 'description':desc})
        if articles:
            feeds.append((section, articles))
        return feeds

    def parse_index(self):
        soup = self.index_to_soup('http://hbr.org/magazine')
        fig = soup.find('figure', attrs={'class': lambda x: x and 'current-issue' in x.split()})
        url = 'https://hbr.org' + fig.find('a', href=True)['href']
        img = fig.find('img')
        self.cover_url = 'https://hbr.org' + img['src']
        self.timefmt = img['alt']
        return self.hbr_parse_toc(url)