What is the difference between dialects? Is "Standard SQL" not enough?

Every vendor added their own extensions. **PostgreSQL** has `RETURNING`, `::` casts, and `WITH RECURSIVE` CTEs. **MySQL** has `LIMIT offset, count` and backtick identifiers. **MSSQL (T-SQL)** uses `TOP N` instead of `LIMIT` and `[square brackets]` for identifiers. **BigQuery** uses `STRUCT` and `ARRAY`. The dialect choice affects how the formatter treats these constructs.

What does "comma position: before" mean?

Classic SQL puts the comma at the end of a line: `id,\n name,\n email`. The **"before" variant** puts the comma at the start of the next line: `id\n , name\n , email`. The win: adding or removing the last column does not touch the previous line, so the git diff is cleaner. Takes some getting used to, but many database engineers swear by it.

What does the Minify mode do?

It **strips comments** (both `-- line` and `/* block */`), **joins all lines** into one, and **collapses repeated whitespace** to a single space. The result is the shortest valid version of the query. Use it when you paste SQL into a string literal in app code, or into a single-line log.

Does it keep my comments?

In **Format** mode, yes. `--` and `/* */` comments survive and the formatter tries to fit them sensibly into the new layout. In **Minify** mode, comments are removed along with the whitespace.

Why does my formatted SQL have weird newlines inside a string literal?

The formatter **never modifies content inside quotes**. If you see newlines in a string, they were already there. That is a signal your original SQL had a multi-line string (sometimes intentional, sometimes a copy-paste accident).

What do I do when the formatter throws a syntax error?

First: check the **dialect**. If you use `LIMIT 10 OFFSET 5` that is PostgreSQL/MySQL/SQLite, not MSSQL (T-SQL uses `OFFSET ... ROWS FETCH NEXT ... ROWS ONLY`). Second: trim sections progressively to find the offending line. Often it contains an unclosed quote or a stray `;` inside a string.

Does it handle procedures, triggers, and functions?

Yes, but **support depends on the dialect**. PL/SQL (Oracle) and T-SQL (MSSQL) are decently supported. PostgreSQL [[PL/pgSQL||an extension for writing functions in Postgres]] formats less well because the function body is embedded as a string. In practice: format the outer shell here, then fix the function body by hand.

I have thousands of lines of SQL. Will it choke?

A file with 10k-20k lines formats in a fraction of a second. Really huge files (hundreds of thousands of lines) may take a moment, because everything is synchronous and blocks the browser thread. Practical limit: **tens of thousands of lines** per batch.

Can I use this formatter in my CI/CD as a SQL style check?

Not directly from this page. Under the hood we use the **`sql-formatter`** npm package, which you can add to your project (`npm install sql-formatter`) and run in a pre-commit hook or GitHub Actions. The result is exactly the same, just local.

SQL Formatter - free | YourDevTools

Dialect

Keywords

Commas

Indent

Blank line between queriesTwo blank lines between separate queries

Input SQL1 lines / 222 B

Formatted SQL

20 lines / 248 B

SELECT
  u.id,
  u.name,
  u.email,
  count(o.id) AS orders
FROM
  users u
  LEFT JOIN orders o ON o.user_id = u.id
WHERE
  u.created_at > '2025-01-01'
GROUP BY
  u.id,
  u.name,
  u.email
HAVING
  count(o.id) > 0
ORDER BY
  orders DESC
LIMIT
  50;

Input (lines)

Output (lines)

Input (size)

222 B

Output (size)

248 B

What does a SQL formatter do?

You paste a long, unformatted SQL query. You get a clean, readable query with indentation, uppercase keywords, and a sensible column layout. Same idea as Prettier for JavaScript, but for SQL.

You pick the dialect (PostgreSQL, MySQL, SQLite, MSSQL, BigQuery, MariaDB, Oracle), because each database has its own keywords and small syntax quirks. You can set keyword case (UPPER vs lower), comma position (after each column, or at the start of the next line), and indent width.

The Minify mode does the opposite: strips comments and folds everything into one line. Handy when you paste a query into a log or a string literal in code.

How to use it

Paste your SQL query into the left panel. It can be one line, no indentation, mixed case, anything.

Pick the dialect of your database from the dropdown. For PostgreSQL pick PostgreSQL, for MySQL/MariaDB pick the matching variant.

Choose keyword case: `UPPER` (classic), `lower` (popular with modern teams), or `preserve` (leave as-is).

Choose comma position: `after` (comma at end of line, classic) or `before` (at the start of the next line, easier to add/remove columns without breaking diffs).

Copy the result with the Copy button. Paste into your IDE, migration tool, or documentation.

The Minify mode collapses the query to a single line. Use it when you want to paste SQL into a string literal in code, or into a one-line log.

When this is useful

Typical situations where a SQL formatter saves time:

Code review where a coworker pasted a query in one line. Paste it here, format it, read it calmly.
Database migration with dozens of CREATE TABLE and ALTER statements. After formatting, the file is ready for a clean PR.
Debugging ORM queries (Prisma, TypeORM, Drizzle). ORMs emit SQL in one long line. The formatter shows what actually hit the database.
Technical documentation, reports, presentations. Formatted SQL looks professional.
Porting SQL between dialects. First format in one dialect, then use our SQL dialect converter to swap AUTO_INCREMENT for SERIAL and similar differences.
Minify for logs or for string literals in code: one long line, no comments, ready to paste into `db.query('...')` in Node.js or Python.

Got a CREATE TABLE and want a Prisma or Drizzle schema? Use our SQL to ORM converter. Working with DBML (Database Markup Language)? See DBML to SQL.

Questions and answers

No. The entire formatter runs in your browser in JavaScript. Nothing hits a server. You can even disconnect from the internet after the page loads and the formatter will still work.

SELECT u.id, u.name, u.email, count(o.id) AS orders FROM users u LEFT JOIN orders o ON o.user_id = u.id WHERE u.created_at > '2025-01-01' GROUP BY u.id, u.name, u.email HAVING count(o.id) > 0 ORDER BY orders DESC LIMIT 50;

What does a SQL formatter do?

You paste a long, unformatted SQL query. You get a clean, readable query with indentation, uppercase keywords, and a sensible column layout. Same idea as Prettier for JavaScript, but for SQL.

The Minify mode does the opposite: strips comments and folds everything into one line. Handy when you paste a query into a log or a string literal in code.

How to use it

Paste your SQL query into the left panel. It can be one line, no indentation, mixed case, anything.

Pick the dialect of your database from the dropdown. For PostgreSQL pick PostgreSQL, for MySQL/MariaDB pick the matching variant.

Choose keyword case: `UPPER` (classic), `lower` (popular with modern teams), or `preserve` (leave as-is).

Choose comma position: `after` (comma at end of line, classic) or `before` (at the start of the next line, easier to add/remove columns without breaking diffs).

Copy the result with the Copy button. Paste into your IDE, migration tool, or documentation.

The Minify mode collapses the query to a single line. Use it when you want to paste SQL into a string literal in code, or into a one-line log.

When this is useful

Typical situations where a SQL formatter saves time:

Code review where a coworker pasted a query in one line. Paste it here, format it, read it calmly.
Database migration with dozens of CREATE TABLE and ALTER statements. After formatting, the file is ready for a clean PR.
Debugging ORM queries (Prisma, TypeORM, Drizzle). ORMs emit SQL in one long line. The formatter shows what actually hit the database.
Technical documentation, reports, presentations. Formatted SQL looks professional.
Porting SQL between dialects. First format in one dialect, then use our SQL dialect converter to swap AUTO_INCREMENT for SERIAL and similar differences.
Minify for logs or for string literals in code: one long line, no comments, ready to paste into `db.query('...')` in Node.js or Python.

Got a CREATE TABLE and want a Prisma or Drizzle schema? Use our SQL to ORM converter. Working with DBML (Database Markup Language)? See DBML to SQL.

Questions and answers

No. The entire formatter runs in your browser in JavaScript. Nothing hits a server. You can even disconnect from the internet after the page loads and the formatter will still work.

SQL Formatter

What does a SQL formatter do?

How to use it

When this is useful

Questions and answers

Related tools

SQL Dialect Converter

SQL to ORM Converter

DBML to SQL Converter

Database Connection String Parser

SQL Formatter

What does a SQL formatter do?

How to use it

When this is useful

Questions and answers

Related tools

SQL Dialect Converter

SQL to ORM Converter

DBML to SQL Converter

Database Connection String Parser