from scrapy.selector import
 
 HtmlXPathSelector
 
def parse(self, response):
	hxs = HtmlXPathSelector(response)
	items = []
 
	newurls = hxs.select('//a/@href').extract()
	validurls = []
        for url in newurls:
                #判斷URL是否合法
                if true: 
                        validurls.append(url)
 
        items.extend([self
 
.make_requests_from_url(url).replace(callback=self.parse) for url in validurls])
 
        sites = hxs.select('//ul/li')
        items = []
        for site in sites:
                item = DmozItem()
                item['title'] = site.select('a/text()').extract()
                item['link'] = site.select('a/@href').extract()
                item['desc'] = site.select('text()').extract()
                items.append(item)
 
        return items