jianshu_hot.py 文件源码

def parse_item(self, response):
        title = response.xpath('//h1[@class="title"]/text()').extract()[0]
        body = response.xpath('//div[@class="show-content"]').extract()[0]
        attr = response.xpath('//script[@data-name="note"]/text()').extract()
        images = response.xpath('//div[@class="image-package"]/img/@src').extract()
        notes = json.loads(attr[0].strip())

        # ??markdown ??
        h = html2text.HTML2Text()
        h.ignore_links = False
        h.inline_links = False
        content = h.handle(body)

        item = JianshuItem()
        item["title"] = title
        item["content"] = content.replace('-\n', '-').replace('\n?', '?')
        item["url"] = notes['url']
        item["slug"] = notes['slug']
        item["views_count"] = notes['views_count']
        item["likes_count"] = notes['likes_count']
        item["images"] = images
        yield item