XPath Cheatsheet for How to Select Nodes#
Absolute Expressions#
Use absolute XPath expressions to target a single node:
| XPath | Description |
|---|---|
/html/head/title | Get the page title node. |
/html/body | Get the <body> element. |
Relative Expressions#
Use relative XPath expressions to target multiple nodes:
| XPath | Description |
|---|---|
//img | Get all <img> image nodes. |
//h1 | Get all <h1> headline nodes. |
Indexing#
Combine relative XPath expressions with indexing to target specific nodes:
| XPath | Description |
|---|---|
//ul[1]/li[3] | Get the third <li> child node of the first <ul> list. |
//ul[1]/li[position()>1] | Get all <li> child nodes of the first <ul> list except the first one. |
//form[@id='login']/input[3] | Get the third <input> child node of a form with a specific id. |
//h2[1]//p | Of the first <h2> headline, get any of its child <p> paragrapds. |
Attributes#
Get nodes with certain attributes with the @ selector:
| XPath | Description |
|---|---|
//div[@class='some_class'] | Get all <div> nodes with with a specific class attribute. |
//*[@id='some_id'] | Get all nodes with the * wildcard selector with a specific id attribute. |
//input[@type="password"] | Get all <input> nodes that are a password type. |
Parents and Children#
Though we most often traverse down the hierachy with / or //, sometimes we need to get parent nodes with ..:
| XPath | Description |
|---|---|
//h1/.. | Get the parent(s) of any <h1> headline nodes. |
//*[@id='some_id']/.. | Get the parent of a node with a specific id. |
//li[1]/../li[9] | Combine parent and children to jump to the ninth list element of the first list. |
See more examples in the axes section.
Union of Several Paths#
Use the pipe | to combine several paths in union:
| XPath | Description |
|---|---|
//h2 | //h3 | Get all <h2> and <h3> headline nodes. |