JSON Minifier

What it does

Minifying removes the whitespace JSON allows between tokens. It changes nothing else, which is the whole property worth checking: parse the output and it must equal the input.

Parse the input, re-serialise with no insignificant whitespace, and report the byte count before and after.

Why it works this way

The whitespace inside a JSON string is data and is preserved. Only the whitespace between tokens is removable, and a minifier that strips inside strings corrupts the payload in a way that still parses.

Minifying is not compression. It removes the whitespace a human added; it does not encode anything. For real size reduction the transport should compress, and this is what is left after that.

The saving is largest on data that was pretty-printed with a wide indent, and zero on JSON that was already on one line — which is why the byte counts are shown rather than a claim about size.

Examples

{
  "a": 1,
  "b": [2, 3]
}

{"a":1,"b":[2,3]}

{"note": "a  b"}

{"note":"a  b"}

Inputs and outputs

• JSON text

• the minified JSON
• bytes before and after

Edge cases

• whitespace inside strings
• already minified — byte counts equal, which is a correct result not a failure
• numbers with exponents and trailing zeros — must not be reformatted

Sources

• RFC 8259, section 2 (insignificant whitespace)