pipelines.py 文件源码

def process_item(self, item, spider):
        if not re.match('.*comment.*',item['link']):
            if re.match('^http.*qq.com.*\.s?html?$',item['link']):
                if item['link'] in self.seen:
                    raise DropItem('Duplicate link %s' % item['link'])
                self.seen.add(item['link'])
                line = json.dumps(dict(item), ensure_ascii=False) + '\n'
                self.file.write(line)
                return item