爬虫数据清洗之html标签删除

删除指定的元素标签

from lxml import etree
import html

def delete_e(d,old_xpath,xpath_list):
    '''
    :param d: 要删除html标签
    :param old_xpath: 最后一级定位标签
    :param xpath_list: 要删除的标签元素集合
    :return: 返回删除后的标签元素
    '''
    content= etree.HTML(d)
    for a in xpath_list:
        try:
            delem=content.xpath(a)[0]
            delem.getparent().remove(delem)
        except:
            pass

    content = content.xpath(old_xpath)[0]
    content = etree.tostring(content, pretty_print=True).decode('utf-8')
    desc = html.unescape(content)
    return desc

例子：
content = response.xpath(’//div[@id=“zoom”]’).extract_first()

   <!-- 假设content的html结构如下：-->
    <div id='zoom'>
        <p> p标签1 </p>
        <p> p标签2 </p>
        <p> p标签3 </p>
        <div> div 标签1 </div>
        <div> div 标签2 </div>
        <span> span标签 </span>
    <di>
   <!--  此时，假设要删除p标签3,使用方法如下：
    只需将xpath定位取出的标签代码赋值给d，将xpath语句的最后一级赋值给old_xpath，
    注意：content的xpath语句只有一级，也就只有一个div，所有不用更改，若xpath语句是 //div[@id="zoom"]/div,则赋值给old_xpath的是'//div'-->
 

content = delete_e(d=content, old_xpath=’//div[@id=“zoom”]’, xpath_list=[’//p[3]’])