XPath语法

XPath的语法:

1. XPath中的符号

符号

说明

示例

示例说明

/

表示从根节点开始选择

/pets

选择根节点pets

表示节点和子节点之间的间隔符

/pets/dog

选择pets节点下的dog节点

//xx

表示从整个xml文档中查找,而不考虑当前节点位置

//price

选择文档中所有的price节点

.

单个英文半角句点表示选择当前节点

/pets/.

选择pets节点

..

双点,表示选择父节点

/pets/dog[0]/..

表示pets节点,也就是第一个dog节点的父节点

@xx

表示选择属性

//dog/@color

表示选择所有dog节点的color属性集合

[…]

中括号表示选择条件,括号内为条件

//dog[@color=’white’]

所有color为white的dog节点

//dog[/price<100]

所有price字节点值小于100的dog节点

中括号内数字为节点索引,类似c#等语言中的数组,数组下标是从1开始的

//dog[1]

第1个dog节点

//dog[last()]

最后一个dog节点,last()是xPath内置函数

|

单竖杠表示合并节点结合

//dog[@color=’white’] | //cat[@color=’white’]

color属性为white的dog节点和color属性为white的cat节点

*

星号表示任何名字的节点或者属性

//dog/*

表示dog节点的所有子节点

//dog/@*

表示dog节点的所有属性节点

2. XPath数学运算符
+  加号表示加
- 表示数字相减
* 表示乘以
div 表示除以,这里数学上的除号/已经被用作节点之间分隔符了
mod 表示取余
 
3. XPath逻辑运算符
= 等于,相当于c#中的 ==
!= 不等于
> 大于
>= 大于等于
< 小于
<= 小于等于
and 并且 与关系
or 或者 或关系
 
 
4. XPath Axes 从字面翻译这个是XPath轴的意思,但根据我的理解这个翻译成XPath节点关系运算关键字更合适,就是一组关键字加上::双冒号表示和当前节点有关系的一个或者一组节点.
使用语法: axisname::nodetest[predicate] 即轴名字::节点名字[取节点条件]
具体说明如下:

关键字

说明

示例

示例说明

ancestor

当前节点的父祖节点

ancestor::pig

当前节点的祖先节点中的pig节点

ancestor-or-self

当前节点以及其父祖节点

ancestor::pig

attribute

当前节点的所有属性

attribute::weight

相当于@weight,attribute::和@是等价的

child

当前节点的所有字节点

child::*[name()!=’price’]

选择名字不是price的子节点

descendant

子孙节点

descendant::*[@*]

有属性的子孙节点

descendant-or-self

子孙节点以及当前节点

descendant-or-self::*

following

Xml文档中当前节点之后的所有节点

following::*

following-sibling

当前节点的同父弟弟节点

following-sibling::

preceding

Xml文档中当前节点之前的所有节点

preceding::*

namespace

选取当前节点的所有命名空间节点

namespace::*

parent

当前节点的父节点

parent::

相当于双点..

preceding-sibling

当前节点之后的同父兄节点

preceding-sibling::*

self

当前节点

self::*

相当于单点.

 
5. 常用的XPath函数介绍:
在XPath表达式中常用的函数有下面两个:
position() 表示节点的序号例如 //cat[position() = 2] 表示取序号为2的dog节点
last() 表示取最后一个节点 //cat[last()] 
name() 表示当前节点名字 /pets/*[name() != 'pig'] 表示/pets下名字不是pig的子节点
 
XPath的函数还有很多,包括字符串函数,数字函数和时间函数等,具体可以参考w3的网站。
 
以上是XPath的语法,下面我们看下如何在.Net中使用XPath
在.Net中可以通过XPathDocument或者XmlDocument类使用XPath。XPathDocument是只读的方式定位Xml节点或者属性文本等,而XmlDocument则是可读写的。
 
如下代码示例展示了如何使用XPathDocument和XmlDocument
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Xml.XPath;
using System.Xml;
 
namespace UseXPathDotNet
{
    class Program
    {
        static void Main(string[] args)
        {
            UseXPathWithXPathDocument();
 
            UseXPathWithXmlDocument();
 
            Console.Read();
        }
 
        static void UseXPathWithXmlDocument()
        {
            XmlDocument doc = new XmlDocument();
            doc.Load("http://www.cnblogs.com/yukaizhao/rss");
            //使用xPath选择需要的节点
            XmlNodeList nodes = doc.SelectNodes("/rss/channel/item[position()<=10]");
            foreach (XmlNode item in nodes)
            {
                string title = item.SelectSingleNode("title").InnerText;
                string url = item.SelectSingleNode("link").InnerText;
                Console.WriteLine("{0} = {1}", title, url);
            }
        }
 
        static void UseXPathWithXPathDocument()
        {
            XPathDocument doc = new XPathDocument("http://www.cnblogs.com/yukaizhao/rss");
            XPathNavigator xPathNav = doc.CreateNavigator();
            //使用xPath取rss中最新的10条随笔
            XPathNodeIterator nodeIterator = xPathNav.Select("/rss/channel/item[position()<=10]");
            while (nodeIterator.MoveNext())
            {
                XPathNavigator itemNav = nodeIterator.Current;
                string title = itemNav.SelectSingleNode("title").Value;
                string url = itemNav.SelectSingleNode("link").Value;
                Console.WriteLine("{0} = {1}",title,url);
            }
 
        }
    }
}

具体参见:

http://www.cnblogs.com/yukaizhao/archive/2011/07/25/xpath.html

http://www.cnblogs.com/loofah/archive/2012/05/10/2494036.html

 

其它有用的示例:

//节点InnerText的等值查询
string xpath = "users/user[username='huo' and password='123']";
//节点InnerText的模糊查询
string xpath = "users/user[contains(username,'huo') and contains(password,'123')]";

//属性的等值查询:
string xpath = "users/user[@username='huo' and @password='123']";
//属性的模糊查询:
string xpath = "users/user[contains(@username,'huo') and contains(@password,'123')]";

如果XPath的开头是一个斜线(/)代表这是绝对路径。如果开头是两个斜线(//)表示文件中所有符合模式的元素都会被选出来,即使是处于树中不同的层级也会被选出来。
以下的语法会选出文件中所有叫做cd的元素(在树中的任何层级都会被选出来):
//cd

选择未知的元素
使用星号(Wildcards,*)可以选择未知的元素。
下面这个语法会选出/catalog/cd 的所有子元素:
/catalog/cd/*
以下的语法会选出所有catalog的子元素中,包含有price作为子元素的元素。
/catalog/*/price
以下的语法会选出有两层父节点,叫做price的所有元素。
/*/*/price
以下的语法会选择出文件中的所有元素。
//*
要注意的是,想要存取不分层级的元素,XPath语法必须以两个斜线开头(//),想要存取未知元素才用星号(*),星号只能代表未知名称的元素,不能代表未知层级的元素。

选择分支
使用中括号可以选择分支。以下的语法从catalog的子元素中取出第一个叫做cd的元素。XPath的定义中没有第0元素这种东西。
/catalog/cd[1]
以下语法选择catalog中的最后一个cd元素:(XPathj并没有定义 first() 这种函式喔,用上例的 [1]就可以取出第一个元素。
/catalog/cd[last()]

原文地址:https://www.cnblogs.com/spriteflk/p/2537027.html