spider.py 文件源码

def parse_item(self, response):
        item = Cl1024Item()
        item['cl_title'] = response.meta['cl_title']
        item['cl_url'] = response.meta['cl_url']
        item['cl_bankuai'] = response.meta['cl_bankuai']
        item['posted'] = response.meta['posted']
        # redownloaded = re.search('downloaded:(.+?)<BR>', response.body)
        # downloaded = redownloaded[12:-4]
        sel = Selector(response)
        downloaded = sel.xpath('//td/table/tr/td/text()').extract()[2]
        item['torrent_downloaded'] = downloaded[17:]
        item['torrent_url'] = response.url
        ref = sel.xpath('//input[@name="ref"]/@value').extract_first()
        reff = sel.xpath('//input[@name="reff"]/@value').extract_first()

        dl = ('http://www.rmdown.com/download.php?ref=%s&&reff=%s&submit=download' % (ref, reff)).encode('utf-8')
        item['torrent_download_urls'] = dl

        yield item