SQL Formatter

How to use SQL Formatter

1. 1
   Paste SQL

   Paste the query you want to format.

2. 2
   Click Format

   Indented, uppercased-keyword SQL appears in the output pane.

About SQL Formatter

MySQL vs PostgreSQL vs SQL Server — the differences that break copy-pasted queries

Formatting is dialect-agnostic, but running the query is not. These are the differences most likely to bite when a query written for one database is pasted into another:

The formatter accepts all of these dialects — the table is about executing the query, not formatting it.

When to use SQL Formatter

• Reviewing ORM output

  ORMs like Prisma and Hibernate produce minified, hard-to-read SQL. Format before reviewing.

• PR code review

  Reformat raw SQL in migrations or seed files before committing for cleaner diffs.

• Learning SQL

  Format example queries from tutorials to see clause structure clearly.

Examples

select u.id, u.name, count(o.id) as orders from users u left join orders o on o.user_id = u.id where u.active = 1 group by u.id, u.name order by orders desc limit 10

SELECT u.id,
  u.name,
  COUNT(o.id) AS orders
FROM users u
LEFT JOIN orders o ON o.user_id = u.id
WHERE u.active = 1
GROUP BY u.id,
  u.name
ORDER BY orders desc
LIMIT 10

SQL mistakes that formatting makes visible

• Double quotes around string literals

  In standard SQL and PostgreSQL, double quotes mean identifiers and single quotes mean strings. MySQL by default tolerates double-quoted strings, so a query that works in MySQL can suddenly mean "compare against a column named active" when pasted into Postgres. Always use single quotes for string values.

  WHERE status = "active"  -- MySQL: string · Postgres: column reference
  WHERE status = 'active'  -- correct everywhere

• Comparing to NULL with =

  NULL is not a value — WHERE col = NULL is never true in any dialect because NULL comparisons yield UNKNOWN, not TRUE. Use IS NULL / IS NOT NULL. The same three-valued logic makes NOT IN return zero rows when its subquery contains a single NULL.

  WHERE email = NULL   -- matches nothing, silently
  WHERE email IS NULL  -- correct

• Accidental Cartesian products hiding in one-liners

  A comma-style FROM users, orders with a forgotten join condition multiplies every row of one table by every row of the other. On a 300-character single line, the missing ON is invisible — after formatting puts each JOIN on its own line, it jumps out. This is the single best reason to format a query before reviewing it.

• Relying on MySQL’s old GROUP BY leniency

  Selecting columns that are neither aggregated nor in GROUP BY errors out in PostgreSQL and in MySQL 5.7+ (ONLY_FULL_GROUP_BY is the default). Queries written against older, lenient MySQL configurations break on migration — aggregate the column or add it to GROUP BY.

Frequently asked questions

Related tools

From the blog