Skip to main content
XPath (XML Path Language) is a powerful query language for selecting nodes from XML and HTML documents. It offers more advanced capabilities than CSS selectors, including text node extraction, parent selection, and complex conditional logic.

​
How XPath Works

XPath uses path expressions to navigate through the hierarchical structure of XML/HTML documents. You can select elements, attributes, text nodes, and more using a syntax similar to file system paths. Key Benefits:
  • Extract text nodes directly (without HTML tags)
  • Navigate to parent elements
  • Use advanced conditional logic
  • Perfect for RSS/XML feeds
  • Support for regex matching

​
XPath Versions in changedetection.io

changedetection.io supports two XPath implementations:
Prefix: xpath: (or no prefix for // syntax)Engine: elementpath libraryFeatures:
  • XPath 2.0 and 3.0 support
  • Better namespace handling
  • Automatic default namespace support for RSS/Atom feeds
  • Modern expression syntax
Example:
xpath://div[@class='price']/text()
//title/text()

​
Basic XPath Syntax

​
Selecting Elements

//div
Selects all <div> elements anywhere in the document. 
/html/body/div
Selects <div> elements that are direct children of <body>.

​
Selecting by Attributes

//div[@class='price']
Selects divs with class="price". 
//a[@href]
Selects all <a> elements that have an href attribute.

​
Extracting Text

//h1/text()
Extracts the text content of <h1> elements (without HTML tags). 
//div[@id='content']//text()
Extracts all text within the content div.

​
Extracting Attributes

//img/@src
Extracts the src attribute from all images. 
//meta[@property='og:price:amount']/@content
Extracts Open Graph price metadata.

​
Practical Examples

​
Monitor Product Price

<div class="product">
  <h2>Gaming Laptop</h2>
  <span class="price" data-value="1299.99">$1,299.99</span>
</div>

​
Extract Multiple Fields

//div[@class='product']/h2/text()
//span[@class='price']/text()
//div[@class='stock']/text()
Each XPath expression on a new line extracts different fields.

​
Monitor RSS Feed Items

Automatic default namespace handling:
//item/title/text()
//item/description/text()
Works directly with RSS feeds without namespace handling.

​
Advanced Techniques

​
Conditional Selection

//div[contains(text(), 'In Stock')]
Selects divs containing “In Stock” text.
//h2[contains(@class, 'product')]/text()
Selects h2 elements where class contains “product”.
//div[@class='product' and @data-available='true']
Selects divs matching BOTH conditions.
//span[@class='sale' or @class='discount']
Selects spans matching EITHER condition.
//div[@class='item' and not(@class='hidden')]
Selects items that are NOT hidden.
//li[1]
Selects the first <li> element.
//li[last()]
Selects the last <li> element.
//li[position() > 2]
Selects all <li> elements after the second one.

​
Parent and Sibling Navigation

//span[@class='price']/parent::div
Selects the parent div of the price span. 
//h2[@class='title']/following-sibling::p[1]
Selects the first paragraph following the title. 
//span[@class='label']/preceding-sibling::input
Selects input elements before the label.

​
Regular Expression Matching

changedetection.io supports EXSLT regex functions: 
//div[re:match(text(), 'Price: \$\d+\.\d{2}')]
Matches divs with text matching the price pattern. 
//a[re:test(@href, 'product/\d+')]
Selects links where href matches the pattern.

​
Working with Namespaces

​
RSS/Atom Feeds

<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
  <channel>
    <title>My Feed</title>
    <item>
      <title>First Item</title>
      <description>Item description</description>
    </item>
  </channel>
</rss>

​
Handling CDATA Sections

Some RSS feeds wrap content in CDATA: 
<description><![CDATA[<p>HTML content here</p>]]></description>
changedetection.io automatically processes CDATA sections. The XPath //description/text() will extract the content.

​
Combining Include and Remove Filters

//article[@class='main']
This extracts the main article while removing ads, sidebars, and scripts.
Remove filters must use the xpath: or xpath1: prefix explicitly.

​
Testing XPath Expressions

​
Using Browser DevTools

  1. Open DevTools Console (F12)
  2. Use $x() to test XPath: 
    $x('//div[@class="price"]/text()')
  3. Verify the results

​
Online XPath Testers

Online testers typically use XPath 1.0. Use xpath1: prefix in changedetection.io for compatibility.

​
Common Patterns

​
Pattern: Extract Plain Text

//div[@id='content']//text()
Use case: Get all text without HTML tags.

​
Pattern: Monitor Table Data

//table[@class='pricing']//tr[2]/td[3]/text()
Use case: Extract specific table cell (row 2, column 3).

​
Pattern: Get Meta Description

//meta[@name='description']/@content
Use case: Extract page meta description.

​
Pattern: Track Stock Status

//div[@class='availability' and contains(text(), 'In Stock')]
Use case: Check if “In Stock” appears. 
//a[@class='download']/@href
Use case: Get download link URLs.

​
Common Pitfalls

Pitfall #1: Forgetting text()
//div[@class='price']
Returns the entire element with HTML tags.Better:
//div[@class='price']/text()
Extracts just the text content.
Pitfall #2: Case SensitivityXPath is case-sensitive!
//Div[@Class='Price']  # Wrong
//div[@class='price']  # Correct
Pitfall #3: Namespace Issues with RSSIf your XPath returns nothing from an RSS feed:Problem: Using XPath 1.0 without local-name()
xpath1://item/title/text()  # May fail
Solution 1: Use default XPath (2.0/3.0)
//item/title/text()  # Automatic namespace handling
Solution 2: Use local-name() with XPath 1.0
xpath1://*[local-name()='item']/*[local-name()='title']/text()

​
When to Use XPath

Good for:
  • Monitoring RSS/Atom feeds
  • Extracting text without HTML tags
  • Complex conditional filtering
  • Navigating to parent elements
  • XML documents
  • When you need regex matching
  • Extracting specific attributes
Not ideal for:
  • JSON APIs (use JSON filtering instead)
  • When CSS selectors are sufficient (simpler syntax)
  • When you want visual selector support

​
XPath vs CSS Selectors

FeatureXPathCSS Selectors
Text extraction//div/text()Not possible
Parent selection//span/parent::divNot possible
Attribute extraction//@hrefNot directly
Conditional logic[contains(@class, 'x')]Limited
Visual selector❌ Not available✅ Available
RSS/XML feeds✅ Excellent❌ Not suitable
Learning curveSteeperEasier

​
Real-World Examples

//div[@itemprop='aggregateRating']//span[@itemprop='ratingValue']/text()
Extracts structured rating data from product pages.
//item/title/text()
//item/pubDate/text()
Monitors RSS feed items for new headlines and dates.
//script[@type='application/ld+json']/text()
Extracts JSON-LD structured data, which can then be filtered with JSON filters.
//table[@class='data']//tr[position() > 1]/td[2]/text()
Extracts second column from all data rows (skipping header).

​
Debugging Tips

​
Returns Empty Results

  1. Check if you’re using the right XPath version
  2. For RSS/XML, try switching between xpath: and xpath1:
  3. Use local-name() for namespaced elements
  4. Verify element exists (check browser’s element inspector)
  5. Test in browser console: $x('your-xpath-here')

​
Returns Unexpected Content

  1. Add /text() to extract only text content
  2. Use [1] or [last()] to get specific positions
  3. Add more specific conditions with [@attribute='value']
  4. Check if you need // (any level) vs / (direct child)