Convert XML API responses to JSON instantly. Fetch any XML endpoint, parse raw XML, and get clean JSON output with attribute handling and array detection.
XML APIs still power a significant portion of enterprise integrations, legacy SOAP services, and data feeds (RSS, Atom, TVDB, weather services). Consuming these in modern JavaScript applications requires conversion to JSON. Manual parsing is error-prone: repeated XML elements must become arrays, attributes need a consistent prefix convention (commonly @attr), and mixed content nodes require careful handling of #text keys. A naive regex-based approach will silently corrupt nested structures or lose attribute data. This tool performs recursive DOM tree walking using the browser's native DOMParser, producing spec-compliant JSON that preserves the full information content of the source XML document.
The converter supports two workflows: fetching a live XML API endpoint (routed through a CORS proxy for browser compatibility) or pasting raw XML directly. It handles CDATA sections, XML namespaces, self-closing tags, and processing instructions. Limitations: XML comments are discarded, and extremely large documents (above 50MB) may exceed browser memory constraints. The attribute prefix and text node key are configurable to match your downstream schema requirements.
The core conversion follows a recursive tree-walk algorithm. For each XML element node E, the converter builds a JSON object J by iterating over attributes and child nodes:
Where prefix is the configurable attribute prefix (default @), textKey is the text content key (default #text). Simplification rule: if J contains only textKey and no attributes, collapse to a plain string value. If the element is empty with no attributes, return null.
| XML Pattern | JSON Output | Rule Applied |
|---|---|---|
<name>John</name> | "name": "John" | Text-only element → string value |
<item>A</item><item>B</item> | "item": ["A", "B"] | Repeated siblings → array |
<node id="5">val</node> | "node": {"@id": "5", "#text": "val"} | Attributes use @ prefix |
<empty/> | "empty": null | Self-closing → null |
<a><b>1</b><c>2</c></a> | "a": {"b": "1", "c": "2"} | Nested elements → nested object |
<![CDATA[raw & data]]> | "#text": "raw & data" | CDATA preserved as text |
<ns:tag xmlns:ns="..."> | "ns:tag": {...} | Namespace prefix retained |
<a attr="1"><b>2</b></a> | "a": {"@attr": "1", "b": "2"} | Mixed attributes + children |
<root>text<b>bold</b>more</root> | "root": {"#text": "textmore", "b": "bold"} | Mixed content merges text nodes |
<val>123</val> | "val": "123" or 123 | Optional: auto-detect numbers |
<flag>true</flag> | "flag": "true" or true | Optional: auto-detect booleans |
<?xml version="1.0"?> | Ignored | Processing instructions stripped |
<!-- comment --> | Ignored | Comments discarded |
<a><b>1</b><b>2</b><c>3</c></a> | "a": {"b": ["1","2"], "c": "3"} | Mixed unique + repeated |
<a> </a> | "a": null (trimmed) | Whitespace-only → null (configurable) |