Why are numbers strings in the JSON?

XML has no number type: 1965 and 0123 are both text. The converter keeps "1965" and "0123" as strings rather than guessing, so an ID or ZIP code never loses a leading zero.

How are XML attributes converted to JSON?

They go into an "@attributes" object on the element, and the element’s text goes under "#text": x gives {"a":{"@attributes":{"id":"1"},"#text":"x"}}. An element with attributes and nothing inside, such as , gives only the "@attributes" object, with no "#text" key. Attributes keep the order they had in the tag.

Why is my XML with a DOCTYPE refused?

A DOCTYPE can define entities that expand into huge text, so any <!DOCTYPE or <!ENTITY is refused and no entity is ever expanded. Only the five predefined entities, & < > " and ', and numeric references such as A are read; © is an error.

What happens to text mixed with child elements?

Hello you there gives "#text": ["Hello","there"] next to "b": "you", with a note. The text pieces keep their order, but where they sat between the child elements is lost, so the JSON cannot rebuild that paragraph exactly.

Can it convert JSON back to XML?

Yes, with JSON → XML. Keys that are not valid XML names are rewritten and listed ("1st" becomes _1st and "ns:item" becomes ns_item), a top-level array or several top-level keys are wrapped in , and control characters XML 1.0 cannot carry are refused with their path.

What happens to comments and namespaces?

Comments and processing instructions are skipped and counted in a note. Namespace prefixes stay in the key names, such as "ns:item" and "xmlns:ns", but are not resolved. Converting that JSON back to XML replaces each colon with an underscore, so v comes back as v : namespaced XML does not survive a round trip.

XML to JSON Converter

Convert XML to JSON and JSON back to XML: attributes go under @attributes, repeated elements become arrays, and an XML error shows its line and column. How it works

Direction:

Format:

Input XML

Up to 2,000,000 characters. Nothing is uploaded.

JSON Output

{
  "library": {
    "book": [
      {
        "@attributes": {
          "id": "b1",
          "lang": "en"
        },
        "title": "Dune",
        "author": "Frank Herbert",
        "year": "1965"
      },
      {
        "@attributes": {
          "id": "b2",
          "lang": "en"
        },
        "title": "Neuromancer",
        "author": "William Gibson",
        "year": "1984"
      }
    ]
  }
}

Skipped 1 comment: comments and processing instructions have no place in the JSON.

9 elements · 4 attributes

Load File

How the mapping works

Every value is text: 0123 stays "0123"; numbers and booleans are not guessed.
Attributes go under "@attributes"; text next to attributes or children goes under "#text"; a repeated element becomes an array.
Comments and processing instructions are skipped (and counted). A DOCTYPE or ENTITY declaration is refused.
Going back, names that are not valid XML names are rewritten and listed, a top-level array is wrapped in <root>, and characters XML 1.0 cannot hold are refused.

About the XML to JSON Converter

Each XML element becomes a JSON key, attributes go under "@attributes", a repeated element becomes an array and every value is a string, so <year>1965</year> gives "year": "1965". The converter also runs the other way, JSON to XML, and reports an XML error with its line and column.

XML and the JSON it gives

XML	JSON (written on one line)
<a>1</a>	{"a":"1"}
<a/>	{"a":""}
<a id="1">x</a>	{"a":{"@attributes":{"id":"1"},"#text":"x"}}
<a><b>1</b><b>2</b></a>	{"a":{"b":["1","2"]}}
<a>0123</a>	{"a":"0123"}
<a><![CDATA[<x>]]></a>	{"a":"<x>"}
<p>Hello <b>you</b> there</p>	{"p":{"b":"you","#text":["Hello","there"]}}

JSON and the XML it gives

JSON	XML (no declaration, on one line)
{"a":{"@attributes":{"id":"1"},"#text":"x"}}	<a id="1">x</a>
{"a":null}	<a/>
[1,2]	<root><item>1</item><item>2</item></root>
{"a":1,"b":2}	<root><a>1</a><b>2</b></root>
{"1st":"x"}	<_1st>x</_1st>
{"ns:item":"v"}	<ns_item>v</ns_item>

XML the converter refuses, and its message

XML	Message
<a><b></a>	Line 1, column 7: The closing tag </a> does not match the open element <b>.
<a>©</a>	Line 1, column 4: The entity "©" is not defined. Only & < > " ' and numeric references such as A are supported (a DOCTYPE cannot define more).
<!DOCTYPE a><a/>	Line 1, column 1: This document has a DOCTYPE / ENTITY declaration. Those are not supported (they can define entities that expand into huge text), so the document is refused. Remove the <!DOCTYPE ...> part.

Frequently asked questions

Why are numbers strings in the JSON?: XML has no number type: <year>1965</year> and <a>0123</a> are both text. The converter keeps "1965" and "0123" as strings rather than guessing, so an ID or ZIP code never loses a leading zero.
How are XML attributes converted to JSON?: They go into an "@attributes" object on the element, and the element’s text goes under "#text": <a id="1">x</a> gives {"a":{"@attributes":{"id":"1"},"#text":"x"}}. An element with attributes and nothing inside, such as <a id="1"/>, gives only the "@attributes" object, with no "#text" key. Attributes keep the order they had in the tag.
Why is my XML with a DOCTYPE refused?: A DOCTYPE can define entities that expand into huge text, so any <!DOCTYPE or <!ENTITY is refused and no entity is ever expanded. Only the five predefined entities, & < > " and ', and numeric references such as A are read; © is an error.
What happens to text mixed with child elements?: <p>Hello <b>you</b> there</p> gives "#text": ["Hello","there"] next to "b": "you", with a note. The text pieces keep their order, but where they sat between the child elements is lost, so the JSON cannot rebuild that paragraph exactly.
Can it convert JSON back to XML?: Yes, with JSON → XML. Keys that are not valid XML names are rewritten and listed ("1st" becomes _1st and "ns:item" becomes ns_item), a top-level array or several top-level keys are wrapped in <root>, and control characters XML 1.0 cannot carry are refused with their path.
What happens to comments and namespaces?: Comments and processing instructions are skipped and counted in a note. Namespace prefixes stay in the key names, such as "ns:item" and "xmlns:ns", but are not resolved. Converting that JSON back to XML replaces each colon with an underscore, so <ns:item xmlns:ns="u">v</ns:item> comes back as <ns_item xmlns_ns="u">v</ns_item>: namespaced XML does not survive a round trip.

Reading an XML feed or API response as JSON

Turn an XML API response, an RSS or Atom feed, a sitemap or a Maven or Android config file into JSON you can query in JavaScript, or turn a JSON object into XML for a system that only takes XML. Paste the text or load a .xml or .json file; the file is read in this browser and its type picks the direction.

How elements and attributes are mapped

The XML is checked for well-formedness by the converter's own reader, then mapped with one convention:

An element with only text becomes that text: <year>1965</year> gives "year": "1965". An empty element gives "".
Attributes go under "@attributes", and text next to attributes or child elements goes under "#text".
A child element that repeats becomes an array in document order.
CDATA counts as text. Comments and processing instructions are skipped and counted.
Every value is a string; numbers and booleans are never guessed.

Example: the sample <library> holds two <book> elements with id and lang attributes. That is 9 elements and 4 attributes, and the JSON has "book" as an array of two objects, each with "@attributes": {"id": "b1", "lang": "en"} or b2, and "year" as a string.

JSON to XML reverses the convention and rewrites keys that are not valid XML names. Limits: 2,000,000 characters and 200 levels of nesting. A DOCTYPE or ENTITY declaration is refused, so no entity is ever expanded.

Related Tools

View all tools

Conversion

CSV to JSON

Convert CSV to JSON: the first row becomes the keys, quoted commas and line breaks are kept, and numbers and booleans are detected only if you ask.

Conversion

JSON to CSV

Turn JSON into CSV: an array of objects, one object or a list of values. Flatten nested objects, expand arrays into rows, then copy or download.

Dev

YAML to JSON Converter

Convert YAML to JSON with anchors and merge keys resolved, YAML 1.2 or 1.1 typing for yes/no, and a warning for numbers JSON cannot hold.

Dev

JSON to YAML

Convert JSON to YAML in block or flow style. Text such as yes or 017 is quoted so it stays a string, and a JSON error shows its line and column.

Dev

SQL to JSON

Turn SQL INSERT statements, or a result table pasted from psql, MySQL or Markdown, into JSON. Quoted values stay strings; numbers, true and NULL get types.

Dev

JSON to SQL

Generate SQL INSERT statements, and CREATE TABLE if you want it, from JSON for PostgreSQL, MySQL, SQLite or SQL Server, with types inferred from every row.