extract.py 文件源码

def select(self):
        search = self.search
        sess = self.login()
        print('Loading search page...')
        for c in range(10):
            sess.visit(self.SEARCH_URI.format(keywords=quote(search)))
            sleep(20 + c)
            sess.interact
            self.client = sess
            if 'results-list' in sess.body():
                break

        soup = BeautifulSoup(sess.body(), 'lxml')
        ul = soup.find('ul', {'class', 'results-list'})
        lis = ul.findAll('li')
        links = dict()
        loop = 5 if len(lis) > 5 else len(lis)
        for idx in range(loop):
            aux = lis[idx].find('div', {'class':'search-result__info'})
            link = aux.find('a', {'class':'search-result__result-link'})
            link = link.get('href') if link else '??'
            name = aux.find('h3', {'class':'search-result__title'})
            name = name.text if name else '??'
            desc = aux.find('p', {'class':'subline-level-1'})
            desc = desc.text if desc else '??'

            if str(idx) not in links.keys():
                links[str(idx)] = dict()
            links[str(idx)].update(dict(link=link, name=name, desc=desc))

        choice = self.printlinks(links)
        return choice