neteasemusicspider.py 文件源码

def parse_follows(self, response):
        ''' parse the follows '''
        url = response.url
        _id = url.split('=')[-1]
        item = response.meta['item']
        driver = response.meta['driver']
        try:
            driver.switch_to.default_content()
            g_iframe = driver.find_elements_by_tag_name('iframe')[0]
            driver.switch_to.frame(g_iframe)
            lis = driver.find_elements_by_xpath('//*[@id="main-box"]/li')
            follows = {}
            for li in lis:
                a = li.find_element_by_tag_name('a')
                title = a.get_attribute('title')
                href = a.get_attribute('href')
                uid = href.split('=')[-1]
                follows[uid] = title
            item['follows'] = follows
        except Exception as e:
            item['follows'] = None
            print e

        # driver.close()
        request = Request(url='http://music.163.com/user/fans?id=' + _id, callback=self.parse_fans)
        request.meta['item'] = copy.deepcopy(item)
        yield request

    # TODO: ??