XML Formatter Online

How the XML Formatter works

XML formatter online: whitespace and attributes

XML does not care about whitespace between elements, and neither do the libraries that generate it, which is why so much of it arrives as a single unbroken line. That is fine for a parser and useless for a person. Formatting puts each element on its own line and indents children under their parent, so the structure of the document — which elements contain which — becomes visible without reading every angle bracket.

Mixed content is kept inline. An element whose only child is text, like <name>Ada</name>, is written on one line rather than being split across three, because breaking it would make simple records three times longer without making them clearer. Elements that contain other elements are expanded, and empty elements are collapsed to the self-closing <tag/> form. An element whose text is mixed with child elements, or spans several lines, is copied exactly as written in both modes, since its spaces and line breaks are part of its content.

Comments, processing instructions and CDATA survive formatting. Each gets its own line at the correct indent, and the contents are never touched — a CDATA block can hold angle brackets and ampersands that would otherwise need escaping, and that is exactly why it exists. Minify drops comments, on the assumption that a minified document is going to a machine, except inside mixed content, which is copied as written. It keeps everything else including the XML declaration and any DOCTYPE.

Validation here means well-formedness, not schema validation. Before anything is reformatted the tool walks the tag structure and checks that every opening tag has a matching closing tag in the right order, that there is exactly one root element, and that no text appears outside it. If that fails you get a message naming the tag it expected and the tag it found, which is usually enough to locate the problem. It does not check the document against an XSD or DTD, does not resolve namespaces, and does not know whether a <price> element should contain a number.

Attributes are normalised lightly. Whitespace around the = in an attribute is removed, and multiple spaces between attributes collapse to one, but quote style is preserved: an attribute written with single quotes stays single-quoted. The order of attributes is never changed, since some tools that consume XML — despite the specification saying they should not — care about it.

Indentation is a choice of convention. Two spaces is the common default for hand-edited configuration, four spaces mirrors most XML produced by IDEs, and tabs let each reader set their own width. None of them changes what the document means. If you need the document as a single line again, Minify produces one with no whitespace between elements, which is the smallest form that still parses identically.

Related tools