登录 注册

php|python|golang|C语言|XMl|前端|数据库|运维|工具类|微信|JAVA|

How to use Python for xpath, JsonPath, and bs4?

1.xpath

1.1 xpath使用

  • google提前安装xpath插件，按ctrl + shift + x 出现小黑框

  • 安装lxml库 pip install lxml ‐i https://pypi.douban.com/simple

  • 导入lxml.etreefrom lxml import etree

  • etree.parse() 解析本地文件html_tree = etree.parse('XX.html')

  • etree.HTML() 服务器响应文件html_tree = etree.HTML(response.read().decode('utf‐8')

  • .html_tree.xpath(xpath路径)

1.2 xpath基本语法

1.路径查询

  • 查找所有子孙节点，不考虑层级关系

  • 找直接子节点

2.谓词查询

//div[@id] 
//div[@id="maincontent"]

3.属性查询

//@class

4.模糊查询

//div[contains(@id, "he")] 
//div[starts‐with(@id, "he")]

5.内容查询

//div/h2/text()

6.逻辑运算

//div[@id="head" and @class="s_down"] 
//title | //price

1.3 示例

xpath.html

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8"/>
    <title>Title</title>
</head>
<body>
    <ul>
        <li id="l1" class="class1">北京</li>
        <li id="l2" class="class2">上海</li>
        <li id="d1">广州</li>
        <li>深圳</li>
    </ul>
</body>
</html>
from lxml import etree

# xpath解析
# 本地文件：                                          etree.parse
# 服务器相应的数据    response.read().decode(&#39;utf-8&#39;)  etree.HTML()


tree = etree.parse(&#39;xpath.html&#39;)

# 查找url下边的li
li_list = tree.xpath(&#39;//body/ul/li&#39;)
print(len(li_list))  # 4

# 获取标签中的内容
li_list = tree.xpath(&#39;//body/ul/li/text()&#39;)
print(li_list)  # [&#39;北京&#39;, &#39;上海&#39;, &#39;广州&#39;, &#39;深圳&#39;]

# 获取带id属性的li
li_list = tree.xpath(&#39;//ul/li[@id]&#39;)
print(len(li_list))  # 3

# 获取id为l1的标签内容
li_list = tree.xpath(&#39;//ul/li[@id="l1"]/text()&#39;)
print(li_list)  # [&#39;北京&#39;]

# 获取id为l1的class属性值
c1 = tree.xpath(&#39;//ul/li[@id="l1"]/@class&#39;)
print(c1)  # [&#39;class1&#39;]

# 获取id中包含l的标签
li_list = tree.xpath(&#39;//ul/li[contains(@id, "l")]/text()&#39;)
print(li_list)  # [&#39;北京&#39;, &#39;上海&#39;]
# 获取id以d开头的标签
li_list = tree.xpath(&#39;//ul/li[starts-with(@id,"d")]/text()&#39;)
print(li_list)  # [&#39;广州&#39;]
# 获取id为l2并且class为class2的标签
li_list = tree.xpath(&#39;//ul/li[@id="l2" and @class="class2"]/text()&#39;)
print(li_list)  # [&#39;上海&#39;]
# 获取id为l2或id为d1的标签
li_list = tree.xpath(&#39;//ul/li[@id="l2"]/text() | //ul/li[@id="d1"]/text()&#39;)
print(li_list)  # [&#39;上海&#39;, &#39;广州&#39;]

1.4 爬取百度搜索按钮的value

import urllib.request
from lxml import etree
url = &#39;http://www.baidu.com&#39;
headers = {
    &#39;User-Agent&#39;: &#39;Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/103.0.0.0 Safari/537.36&#39;
}
request = urllib.request.Request(url=url, headers=headers)
response = urllib.request.urlopen(request)
content = response.read().decode(&#39;utf-8&#39;)
tree = etree.HTML(content)
value = tree.xpath(&#39;//input[@id="su"]/@value&#39;)
print(value)

Python xpath,JsonPath,bs4怎么使用

1.5 爬取站长素材的图片

# 需求 下载的前十页的图片
# https://sc.chinaz.com/tupian/qinglvtupian.html   1
# https://sc.chinaz.com/tupian/qinglvtupian_page.html
import urllib.request
from lxml import etree
def create_request(page):
    if (page == 1):
        url = &#39;https://sc.chinaz.com/tupian/qinglvtupian.html&#39;
    else:
        url = &#39;https://sc.chinaz.com/tupian/qinglvtupian_&#39; + str(page) + &#39;.html&#39;
    headers = {
        &#39;User-Agent&#39;: &#39;Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/92.0.4515.159 Safari/537.36&#39;,
    }
    request = urllib.request.Request(url=url, headers=headers)
    return request
def get_content(request):
    response = urllib.request.urlopen(request)
    content = response.read().decode(&#39;utf-8&#39;)
    return content
def down_load(content):
    #     下载图片
    # urllib.request.urlretrieve(&#39;图片地址&#39;,&#39;文件的名字&#39;)
    tree = etree.HTML(content)
    name_list = tree.xpath(&#39;//div[@id="container"]//a/img/@alt&#39;)
    # 一般设计图片的网站都会进行懒加载
    src_list = tree.xpath(&#39;//div[@id="container"]//a/img/@src2&#39;)
    print(src_list)
    for i in range(len(name_list)):
        name = name_list[i]
        src = src_list[i]
        url = &#39;https:&#39; + src
        urllib.request.urlretrieve(url=url, filename=&#39;./loveImg/&#39; + name + &#39;.jpg&#39;)
if __name__ == &#39;__main__&#39;:
    start_page = int(input(&#39;请输入起始页码&#39;))
    end_page = int(input(&#39;请输入结束页码&#39;))

    for page in range(start_page, end_page + 1):
        # (1) 请求对象的定制
        request = create_request(page)
        # （2）获取网页的源码
        content = get_content(request)
        # （3）下载
        down_load(content)

2. JsonPath

2.1 pip安装

pip install jsonpath

2.2 jsonpath的使用

obj = json.load(open(&#39;json文件&#39;, &#39;r&#39;, encoding=&#39;utf‐8&#39;)) 
ret = jsonpath.jsonpath(obj, &#39;jsonpath语法&#39;)

JSONPath语法元素和对应XPath元素的对比：

Python xpath,JsonPath,bs4怎么使用

示例：

jsonpath.json

{ "store": {
    "book": [
      { "category": "修真",
        "author": "六道",
        "title": "坏蛋是怎样练成的",
        "price": 8.95
      },
      { "category": "修真",
        "author": "天蚕土豆",
        "title": "斗破苍穹",
        "price": 12.99
      },
      { "category": "修真",
        "author": "唐家三少",
        "title": "斗罗大陆",
        "isbn": "0-553-21311-3",
        "price": 8.99
      },
      { "category": "修真",
        "author": "南派三叔",
        "title": "星辰变",
        "isbn": "0-395-19395-8",
        "price": 22.99
      }
    ],
    "bicycle": {
      "author": "老马",
      "color": "黑色",
      "price": 19.95
    }
  }
}
import json
import jsonpath

obj = json.load(open(&#39;jsonpath.json&#39;, &#39;r&#39;, encoding=&#39;utf-8&#39;))

# 书店所有书的作者
author_list = jsonpath.jsonpath(obj, &#39;$.store.book[*].author&#39;)
print(author_list)  # [&#39;六道&#39;, &#39;天蚕土豆&#39;, &#39;唐家三少&#39;, &#39;南派三叔&#39;]

# 所有的作者
author_list = jsonpath.jsonpath(obj, &#39;$..author&#39;)
print(author_list)  # [&#39;六道&#39;, &#39;天蚕土豆&#39;, &#39;唐家三少&#39;, &#39;南派三叔&#39;, &#39;老马&#39;]

# store下面的所有的元素
tag_list = jsonpath.jsonpath(obj, &#39;$.store.*&#39;)
print(
    tag_list)  # [[{&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;六道&#39;, &#39;title&#39;: &#39;坏蛋是怎样练成的&#39;, &#39;price&#39;: 8.95}, {&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;天蚕土豆&#39;, &#39;title&#39;: &#39;斗破苍穹&#39;, &#39;price&#39;: 12.99}, {&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;唐家三少&#39;, &#39;title&#39;: &#39;斗罗大陆&#39;, &#39;isbn&#39;: &#39;0-553-21311-3&#39;, &#39;price&#39;: 8.99}, {&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;南派三叔&#39;, &#39;title&#39;: &#39;星辰变&#39;, &#39;isbn&#39;: &#39;0-395-19395-8&#39;, &#39;price&#39;: 22.99}], {&#39;author&#39;: &#39;老马&#39;, &#39;color&#39;: &#39;黑色&#39;, &#39;price&#39;: 19.95}]

# store里面所有东西的price
price_list = jsonpath.jsonpath(obj, &#39;$.store..price&#39;)
print(price_list)  # [8.95, 12.99, 8.99, 22.99, 19.95]

# 第三个书
book = jsonpath.jsonpath(obj, &#39;$..book[2]&#39;)
print(book)  # [{&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;唐家三少&#39;, &#39;title&#39;: &#39;斗罗大陆&#39;, &#39;isbn&#39;: &#39;0-553-21311-3&#39;, &#39;price&#39;: 8.99}]

# 最后一本书
book = jsonpath.jsonpath(obj, &#39;$..book[(@.length-1)]&#39;)
print(book)  # [{&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;南派三叔&#39;, &#39;title&#39;: &#39;星辰变&#39;, &#39;isbn&#39;: &#39;0-395-19395-8&#39;, &#39;price&#39;: 22.99}]
# 	前面的两本书
book_list = jsonpath.jsonpath(obj, &#39;$..book[0,1]&#39;)
# book_list = jsonpath.jsonpath(obj,&#39;$..book[:2]&#39;)
print(
    book_list)  # [{&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;六道&#39;, &#39;title&#39;: &#39;坏蛋是怎样练成的&#39;, &#39;price&#39;: 8.95}, {&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;天蚕土豆&#39;, &#39;title&#39;: &#39;斗破苍穹&#39;, &#39;price&#39;: 12.99}]

# 条件过滤需要在（）的前面添加一个？
# 	 过滤出所有的包含isbn的书。
book_list = jsonpath.jsonpath(obj, &#39;$..book[?(@.isbn)]&#39;)
print(
    book_list)  # [{&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;唐家三少&#39;, &#39;title&#39;: &#39;斗罗大陆&#39;, &#39;isbn&#39;: &#39;0-553-21311-3&#39;, &#39;price&#39;: 8.99}, {&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;南派三叔&#39;, &#39;title&#39;: &#39;星辰变&#39;, &#39;isbn&#39;: &#39;0-395-19395-8&#39;, &#39;price&#39;: 22.99}]
# 哪本书超过了10块钱
book_list = jsonpath.jsonpath(obj, &#39;$..book[?(@.price>10)]&#39;)
print(
    book_list)  # [{&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;天蚕土豆&#39;, &#39;title&#39;: &#39;斗破苍穹&#39;, &#39;price&#39;: 12.99}, {&#39;category&#39;: &#39;修真&#39;, &#39;author&#39;: &#39;南派三叔&#39;, &#39;title&#39;: &#39;星辰变&#39;, &#39;isbn&#39;: &#39;0-395-19395-8&#39;, &#39;price&#39;: 22.99}]

3. BeautifulSoup

3.1 基本简介

1.安装

pip install bs4

2.导入

from bs4 import BeautifulSoup

3.创建对象

  • 服务器响应的文件生成对象 soup = BeautifulSoup(response.read().decode(), 'lxml')

  • 本地文件生成对象 soup = BeautifulSoup(open('1.html'), 'lxml')

注意：默认打开文件的编码格式gbk所以需要指定打开编码格式utf-8

3.2 安装以及创建

1.根据标签名查找节点 
	soup.a 【注】只能找到第一个a 
		soup.a.name 
		soup.a.attrs 
2.函数 
	(1).find(返回一个对象) 
		find(&#39;a&#39;)：只找到第一个a标签
		find(&#39;a&#39;, title=&#39;名字&#39;) 
		find(&#39;a&#39;, class_=&#39;名字&#39;) 
	(2).find_all(返回一个列表) 
		find_all(&#39;a&#39;) 查找到所有的a 
		find_all([&#39;a&#39;, &#39;span&#39;]) 返回所有的a和span 
		find_all(&#39;a&#39;, limit=2) 只找前两个a 
	(3).select(根据选择器得到节点对象)【推荐】 
		1.element 
			eg:p 
		2..class 
			eg:.firstname 
		3.#id
			eg:#firstname 
		4.属性选择器 
			[attribute] 
				eg:li = soup.select(&#39;li[class]&#39;) 
			[attribute=value] 
				eg:li = soup.select(&#39;li[class="hengheng1"]&#39;) 
		5.层级选择器 
			element element 
				div p 
			element>element 
				div>p 
			element,element 
				div,p 
					eg:soup = soup.select(&#39;a,span&#39;)

3.3 节点定位

1.根据标签名查找节点 
	soup.a 【注】只能找到第一个a 
		soup.a.name 
		soup.a.attrs 
2.函数 
	(1).find(返回一个对象) 
		find(&#39;a&#39;)：只找到第一个a标签
		find(&#39;a&#39;, title=&#39;名字&#39;) 
		find(&#39;a&#39;, class_=&#39;名字&#39;) 
	(2).find_all(返回一个列表) 
		find_all(&#39;a&#39;) 查找到所有的a 
		find_all([&#39;a&#39;, &#39;span&#39;]) 返回所有的a和span 
		find_all(&#39;a&#39;, limit=2) 只找前两个a 
	(3).select(根据选择器得到节点对象)【推荐】 
		1.element 
			eg:p 
		2..class 
			eg:.firstname 
		3.#id
			eg:#firstname 
		4.属性选择器 
			[attribute] 
				eg:li = soup.select(&#39;li[class]&#39;) 
			[attribute=value] 
				eg:li = soup.select(&#39;li[class="hengheng1"]&#39;) 
		5.层级选择器 
			element element 
				div p 
			element>element 
				div>p 
			element,element 
				div,p 
					eg:soup = soup.select(&#39;a,span&#39;)

3.5 节点信息 

(1).获取节点内容：适用于标签中嵌套标签的结构 
	obj.string 
	obj.get_text()【推荐】 
(2).节点的属性 
	tag.name 获取标签名 
		eg:tag = find(&#39;li) 
			print(tag.name) 
	tag.attrs将属性值作为一个字典返回 
(3).获取节点属性 
	obj.attrs.get(&#39;title&#39;)【常用】 
	obj.get(&#39;title&#39;) 
	obj[&#39;title&#39;]
(1).获取节点内容：适用于标签中嵌套标签的结构 
	obj.string 
	obj.get_text()【推荐】 
(2).节点的属性 
	tag.name 获取标签名 
		eg:tag = find(&#39;li) 
			print(tag.name) 
	tag.attrs将属性值作为一个字典返回 
(3).获取节点属性 
	obj.attrs.get(&#39;title&#39;)【常用】 
	obj.get(&#39;title&#39;) 
	obj[&#39;title&#39;]

3.6 使用示例

bs4.html

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <title>Title</title>
</head>
<body>

    <div>
        <ul>
            <li id="l1">张三</li>
            <li id="l2">李四</li>
            <li>王五</li>
            <a href="" id=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" " class="a1">google</a>
            <span>嘿嘿嘿</span>
        </ul>
    </div>


    <a href="" title=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" a2">百度</a>

    <div id="d1">
        <span>
            哈哈哈
        </span>
    </div>

    <p id="p1" class="p1">呵呵呵</p>
</body>
</html>
from bs4 import BeautifulSoup
# 通过解析本地文件 来将bs4的基础语法进行讲解
# 默认打开的文件的编码格式是gbk 所以在打开文件的时候需要指定编码
soup = BeautifulSoup(open(&#39;bs4.html&#39;, encoding=&#39;utf-8&#39;), &#39;lxml&#39;)
# 根据标签名查找节点
# 找到的是第一个符合条件的数据
print(soup.a)  # <a class="a1" href="" id=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" ">google</a>
# 获取标签的属性和属性值
print(soup.a.attrs)  # {&#39;href&#39;: &#39;&#39;, &#39;id&#39;: &#39;&#39;, &#39;class&#39;: [&#39;a1&#39;]}
# bs4的一些函数
# （1）find
# 返回的是第一个符合条件的数据
print(soup.find(&#39;a&#39;))  # <a class="a1" href="" id=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" ">google</a>
# 根据title的值来找到对应的标签对象
print(soup.find(&#39;a&#39;, title="a2"))  # <a href="" title=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" a2">百度</a>

# 根据class的值来找到对应的标签对象  注意的是class需要添加下划线
print(soup.find(&#39;a&#39;, class_="a1"))  # <a class="a1" href="" id=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" ">google</a>

# （2）find_all  返回的是一个列表 并且返回了所有的a标签
print(soup.find_all(&#39;a&#39;))  # [<a class="a1" href="" id=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" ">google</a>, <a href="" title=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" a2">百度</a>]

# 如果想获取的是多个标签的数据 那么需要在find_all的参数中添加的是列表的数据
print(soup.find_all([&#39;a&#39;,&#39;span&#39;]))  # [<a class="a1" href="" id=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" ">google</a>, <span>嘿嘿嘿</span>, <a href="" title=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" a2">百</a><spa哈</span>]

# limit的作用是查找前几个数据
print(soup.find_all(&#39;li&#39;, limit=2))  # [<li id="l1">张三</li>, <li id="l2">李四</li>]

# （3）select（推荐）
# select方法返回的是一个列表  并且会返回多个数据
print(soup.select(&#39;a&#39;))  # [<a class="a1" href="" id=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" ">google</a>, <a href="" title=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" a2">百度</a>]

# 可以通过.代表class  我们把这种操作叫做类选择器
print(soup.select(&#39;.a1&#39;))  # [<a class="a1" href="" id=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" ">google</a>]

print(soup.select(&#39;#l1&#39;))  # [<li id="l1">张三</li>]

# 属性选择器---通过属性来寻找对应的标签
# 查找到li标签中有id的标签
print(soup.select(&#39;li[id]&#39;))  # [<li id="l1">张三</li>, <li id="l2">李四</li>]

# 查找到li标签中id为l2的标签
print(soup.select(&#39;li[id="l2"]&#39;))  # [<li id="l2">李四</li>]

# 层级选择器
#  后代选择器
# 找到的是div下面的li
print(soup.select(&#39;div li&#39;))  # [<li id="l1">张三</li>, <li id="l2">李四</li>, <li>王五</li>]

# 子代选择器
#  某标签的第一级子标签
# 注意：很多的计算机编程语言中 如果不加空格不会输出内容  但是在bs4中 不会报错 会显示内容
print(soup.select(&#39;div > ul > li&#39;))  # [<li id="l1">张三</li>, <li id="l2">李四</li>, <li>王五</li>]

# 找到a标签和li标签的所有的对象
print(soup.select(
    &#39;a,li&#39;))  # [<li id="l1">张三</li>, <li id="l2">李四</li>, <li>王五</li>, <a class="a1" href="" id=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" ">google</a>, <a href="" title=" rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow"  rel="external nofollow" a2">百度</a>]

# 节点信息
#    获取节点内容
obj = soup.select(&#39;#d1&#39;)[0]
# 如果标签对象中 只有内容 那么string和get_text()都可以使用
# 如果标签对象中 除了内容还有标签 那么string就获取不到数据 而get_text()是可以获取数据
# 我们一般情况下  推荐使用get_text()
print(obj.string)  # None
print(obj.get_text())  # 哈哈哈

# 节点的属性
obj = soup.select(&#39;#p1&#39;)[0]
# name是标签的名字
print(obj.name)  # p
# 将属性值左右一个字典返回
print(obj.attrs)  # {&#39;id&#39;: &#39;p1&#39;, &#39;class&#39;: [&#39;p1&#39;]}

# 获取节点的属性
obj = soup.select(&#39;#p1&#39;)[0]
#
print(obj.attrs.get(&#39;class&#39;))  # [&#39;p1&#39;]
print(obj.get(&#39;class&#39;))  # [&#39;p1&#39;]
print(obj[&#39;class&#39;])  # [&#39;p1&#39;]

3.7 解析星巴克产品名称

import urllib.request
url = &#39;https://www.starbucks.com.cn/menu/&#39;
response = urllib.request.urlopen(url)
content = response.read().decode(&#39;utf-8&#39;)
from bs4 import BeautifulSoup
soup = BeautifulSoup(content,&#39;lxml&#39;)
# //ul[@class="grid padded-3 product"]//strong/text()
# 一般先用xpath方式通过google插件写好解析的表达式
name_list = soup.select(&#39;ul[class="grid padded-3 product"] strong&#39;)
for name in name_list:
    print(name.get_text())

Python xpath,JsonPath,bs4怎么使用

以上就是How to use Python for xpath, JsonPath, and bs4?的详细内容，更多请关注其它相关文章！