XML CDATA Sections

What Is a CDATA Section?

CDATA stands for Character Data. A CDATA section is a special XML construct that tells the parser to treat everything inside it as raw text — not as markup. This lets you include characters like < and & without escaping them.

Normally, these characters must be written as < and &. Inside a CDATA section, you can write them literally:

<description><![CDATA[Price: <$10 & free shipping]]></description>

The parser delivers the text content as Price: <$10 & free shipping — exactly as written, no escaping needed.

CDATA Syntax

A CDATA section has three parts:

<![CDATA[  ...content...  ]]>

• Opening delimiter: <![CDATA[
• Content: any text except the closing delimiter
• Closing delimiter: ]]>

CDATA sections can appear inside element content — never inside a tag name, attribute, or processing instruction. A CDATA section is equivalent to the escaped version of the same text; the XML data model treats them identically.

When to Use CDATA vs Escaping

Both approaches produce identical parsed output. Choose based on readability and context:

Use CDATA when the content contains many special characters

Embedding HTML snippets, SQL queries, code fragments, or regex patterns in XML is much cleaner in a CDATA section than escaping every <, >, and &.

Use escaping when the content has few special characters

A single ampersand in a company name (AT&T) is cleaner escaped than wrapped in a CDATA section.

Use escaping in attribute values

CDATA sections cannot appear inside attribute values — only entity references work there: title="a & b".

The CDATA Nesting Gotcha

CDATA sections cannot be nested, and the closing delimiter ]]> cannot appear inside a CDATA section. If your content contains the string ]]>, you must split the section:

<!-- INVALID: ]]> inside CDATA terminates it early -->
<![CDATA[a]]>b]]>

<!-- Valid: split at the delimiter -->
<![CDATA[a]]]]><![CDATA[>b]]>

This edge case is rare in practice — it only occurs when embedding CDATA-containing XML inside another XML document.

CDATA in Real-World XML

Common places you will encounter CDATA sections:

• SOAP envelopes: message bodies sometimes embed HTML or escaped payloads in CDATA
• RSS/Atom feeds: <description><![CDATA[...HTML...]]></description> is the standard pattern for HTML content in feed items
• Android strings.xml: strings containing apostrophes or HTML tags often use CDATA
• Maven pom.xml: plugin configurations occasionally embed shell scripts in CDATA sections
• SVG: embedded script or style content in SVG files sometimes uses CDATA

How This Formatter Handles CDATA

The XML Formatter preserves CDATA sections during Format and Minify operations. The content between <![CDATA[ and ]]> is never modified or re-escaped — only surrounding whitespace outside the section may be adjusted during formatting.

The Validate function treats CDATA sections as plain text and does not report them as errors. The XML→JSON conversion delivers the raw text content of a CDATA section as the string value of the corresponding JSON key — the CDATA wrapper is transparent to the caller.