Convert XML to JSON Online

XML allows both child elements and attributes to store data, with no direct JSON equivalent. The most widely adopted convention (used by xml2js in Node.js, xmltodict in Python, and JAXB in Java) is to prefix attributes with @: the element <user id="42" role="admin"> becomes {"user": {"@id": "42", "@role": "admin"}}. Text content of mixed elements (those with both attributes and text) is placed in the #text key: <title lang="en">Hello</title> becomes {"title": {"@lang": "en", "#text": "Hello"}}. This convention is consistent with the BadgerFish specification and the most popular ecosystem tools.

In XML, it is perfectly valid to have multiple sibling elements with the same name: <items><item>A</item><item>B</item></items>. JSON cannot represent multiple identical keys in the same object (that would be invalid JSON per RFC 7159). The standard solution is to group them into an array: {"items": {"item": ["A", "B"]}}. A good converter automatically detects repeated elements and converts them to arrays, even when there are only two. This can cause inconsistencies if sometimes there is one element and sometimes several: the JSON would have a string in one case and an array in another. The solution is to always normalize to array for fields that can repeat.

XML namespaces (defined by the W3C in 'Namespaces in XML 1.0', published January 1999 and revised August 2006) are an XML feature with no JSON equivalent. The most common representation is to include the namespace prefix as part of the key name: the element <soap:Body xmlns:soap="http://schemas.xmlsoap.org/soap/envelope/"> becomes {"soap:Body": {...}}. Namespace declarations (xmlns:prefix="uri") are treated as attributes and appear as @xmlns:prefix. For use cases like SOAP-to-REST migration where namespaces are pervasive, post-processing is often needed to strip namespace prefixes from the resulting JSON.

CDATA (Character Data) is a special XML section that allows including text containing normally reserved characters (< > & ' ") without using XML entities (&lt; &gt; &amp;). The syntax is <![CDATA[content here]]>. CDATA is frequently used to include HTML, JavaScript, SQL, or any text with special characters inside an XML element. In JSON conversion, CDATA content is treated as plain text: CDATA sections are replaced by their content without the markup, and the result is placed as a normal string in the JSON field value. If XML has <description><![CDATA[<b>Bold text</b>]]></description>, the resulting JSON is {"description": "<b>Bold text</b>"}.

Conversion happens in your browser using the browser's native DOMParser to parse XML and a custom JSON serializer. DOMParser can handle XML files of several tens of MB in modern browsers (Chrome, Firefox, Edge). For large files (over 10 MB), the process may take several seconds. For very large XML files (over 100 MB, like sitemaps with hundreds of thousands of URLs or XML database dumps), we recommend streaming tools like xml2json in Node.js with streams, xmltodict in Python with xml.etree's iterparse mode, or the Saxon processor for enterprise-grade XML.