UUID Regex

What is a UUID Regex

There are two useful UUID patterns, and picking the wrong one is the usual reason a validator lets bad data through. The loose pattern only demands 32 hex digits in the 8-4-4-4-12 layout — it accepts a version nibble that no RFC defines. The strict pattern also constrains the version digit to 1-8 and the variant nibble to [89ab], which is what RFC 9562 actually specifies.

Both need anchors (^ and $) and the case-insensitive flag, otherwise a valid UUID embedded in a longer string will match and a truncated value in the middle of a sentence will look fine. Copy the pattern for your language below, or paste a value into the tester to see which pattern it satisfies.

How to

1. Choose your language, then copy the pattern — each one is anchored and case-insensitive.
2. Or paste values into the field on the left: each line is tested against the strict pattern.
3. Use the table to see the detected version, which tells you whether the value is random (v4) or time-based (v7).

Use cases

• Input validation in a form or API handler, where a bad identifier should be rejected before it reaches the database.
• Log and CSV parsing, pulling identifiers out of text without accidentally matching a 32-character hash.
• Database constraints, when a column has to stay text but must never contain anything but a UUID.

Compare UUID Regex with other formats

Code examples

JavaScript

// Strict, anchored, case-insensitive
const UUID = /^[0-9a-f]{8}-[0-9a-f]{4}-[1-8][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$/i;

// Loose: layout only
const LOOSE = /^[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}$/i;

// Grab every UUID inside a longer string
'ids ' + '550e8400-e29b-41d4-a716-446655440000'.repeat(2)
  .match(/[0-9a-f]{8}-[0-9a-f]{4}-[1-8][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}/gi);

Python

import re

STRICT = re.compile(r'^[0-9a-f]{8}-[0-9a-f]{4}-[1-8][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$', re.I)
LOOSE  = re.compile(r'^[0-9a-f]{8}(-?[0-9a-f]{4}){3}-?[0-9a-f]{12}$', re.I)

bool(STRICT.match('550e8400-e29b-41d4-a716-446655440000'))   # True
bool(STRICT.match('550e8400-e29b-01d4-a716-446655440000'))   # False

Java

import java.util.regex.Pattern;

Pattern STRICT = Pattern.compile(
    "^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-" +
    "[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$");

STRICT.matcher(value).matches();   // true / false

PHP

<?php
$strict = '/^[0-9a-f]{8}-[0-9a-f]{4}-[1-8][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$/i';

if (preg_match($strict, $value) === 1) {
    // well-formed UUID
}

Go

package main

import "regexp"

var strict = regexp.MustCompile(`(?i)^[0-9a-f]{8}-[0-9a-f]{4}-[1-8][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$`)

strict.MatchString(value)

MySQL

-- MySQL 8 supports REGEXP in a CHECK constraint
ALTER TABLE items ADD CONSTRAINT chk_uuid CHECK (
  id REGEXP '^[0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}$'
);

SELECT UUID();            -- v1
SELECT UUID_TO_BIN(UUID(), 1);   -- v1 bytes, time part swapped for sorting

PostgreSQL

-- Regex input check
SELECT '550e8400-e29b-41d4-a716-446655440000'
       ~ '^[0-9a-f]{8}-[0-9a-f]{4}-[1-8][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$';

-- Better: use the native type and let PostgreSQL 18 report the details
SELECT uuid_extract_version('018f3b7a-9c2d-7f41-a1b2-c3d4e5f60718');   -- 7

Frequently asked questions