XML Formatter and Validator

It parses the document, then writes it back with one element per line and each level indented. A few things are kept exactly as written, because changing them would change the data:

• Text and entities. &, © and your own entities stay as entities. Only the whitespace used for indentation between tags is replaced.
• Mixed content. When an element holds text and tags together, like <p>Hello <b>world</b></p>, it stays on one line so no spaces are added to the text.
• xml:space="preserve". The XML spec uses this attribute to say whitespace inside an element matters. Those elements are left untouched.
• CDATA, comments, the declaration and DOCTYPE. All kept. Turn on Remove comments to drop comments.

Empty elements are written as <a/> by default; untick the option to get <a></a>. Both mean the same thing to a parser.

These are the rules from the XML 1.0 spec that most often break a document, with the message you see here.

XML has five built-in entities: <, >, &, ' and ". HTML names like   are not defined in XML unless a DTD declares them; use the number instead ( ).

Command line (libxml2). xmllint ships with macOS and most Linux distributions.

xmllint --format input.xml > pretty.xml   # indent
xmllint --noout input.xml                  # check only; prints errors with line numbers
XMLLINT_INDENT=$'\t' xmllint --format input.xml   # indent with tabs

Python. ElementTree.indent() was added in Python 3.9.

import xml.etree.ElementTree as ET

tree = ET.parse("input.xml")          # raises ET.ParseError with line and column
ET.indent(tree, space="  ")
tree.write("pretty.xml", encoding="utf-8", xml_declaration=True)

JavaScript in the browser. DOMParser does not throw on bad XML; it returns a document that contains a <parsererror> element.

const doc = new DOMParser().parseFromString(xml, "application/xml");
const err = doc.querySelector("parsererror");
if (err) console.error(err.textContent);

Node.js. fast-xml-parser, the library this page uses, has a validator that reports line and column.

import { XMLValidator } from "fast-xml-parser";

const result = XMLValidator.validate(xml);
if (result !== true) console.log(result.err.line, result.err.col, result.err.msg);

The Tree tab shows the document as nested elements with their attributes. The first two levels are open; click the arrow to fold or unfold, or use Expand all. A collapsed element shows how many child elements it has in braces.

Hover a row and click [xpath] to copy a path that selects exactly that element, such as /orders/order[2]/customer/name. The index is added only where an element has siblings with the same name. Paste it into an XPath tester, xmllint --xpath or document.evaluate().

Need the data as JSON instead? The XML to JSON converter turns the same document into JSON with your choice of attribute and array handling. For JSON itself, use the JSON formatter.