Developer Tools
Data & AnalyticsSQL Formatter
Beautify and format raw SQL queries with configurable indentation, keyword casing, and dialect support for MySQL, PostgreSQL, and T-SQL.
Formatting Options
SQL Input
Formatted SQL
Formatted SQL will appear here…Related Tools
About SQL Formatter
How It Works
- Tokenizes your SQL into keywords, identifiers, operators, and literals
- Re-indents clause keywords (SELECT, FROM, WHERE, JOIN, etc.) onto new lines
- Preserves quoted strings, comments, and identifiers exactly as written
- Applies your chosen keyword casing (UPPER, lower, or preserve)
- Supports Standard SQL, MySQL, PostgreSQL, and T-SQL dialects
Common Use Cases
- Making auto-generated ORM queries readable for debugging
- Reviewing query plans and execution logic in code reviews
- Preparing SQL for documentation or sharing with teammates
- Normalizing mixed-case or minified SQL from log files
- Cleaning up machine-generated migration scripts
Frequently Asked Questions
What SQL dialects does the formatter support?
The formatter supports Standard SQL, MySQL, PostgreSQL, and T-SQL (SQL Server). Selecting a dialect adjusts identifier quoting hints and dialect-specific keyword handling so the output is consistent with the target database engine.
Does formatting change the meaning of my query?
No. The formatter only changes whitespace and keyword casing — it never rewrites, reorders, or removes any part of your query. The formatted output is semantically identical to the original input.
What is comma-first style?
Comma-first (also called leading-comma) style places the comma at the beginning of each new item in a list rather than at the end of the previous one. Some teams prefer this because it makes it easier to spot missing commas in version-control diffs.
Why should I use UPPERCASE for SQL keywords?
Using UPPERCASE for reserved words (SELECT, FROM, WHERE, etc.) and lowercase for identifiers (table and column names) is the most widely adopted SQL style convention. It improves readability and immediately distinguishes language keywords from user-defined names.
Can I format multiple SQL statements at once?
Yes. Paste multiple statements separated by semicolons and the formatter will handle each one. Each semicolon is preserved at the end of its statement in the output.
Are single-quoted strings and comments preserved exactly?
Yes. The tokenizer treats single-quoted strings, double-quoted identifiers, backtick identifiers, bracket identifiers, and both -- single-line and /* block */ comments as opaque tokens. Their content is never modified.
Does the tool send my SQL to a server?
No. All formatting is performed entirely in your browser using JavaScript. Your SQL queries are never transmitted to any server, so sensitive schema names, column names, and data literals remain completely private.
What indentation options are available?
You can choose between 2 spaces, 4 spaces, or a tab character for indentation. 2-space indentation is common in JavaScript-adjacent projects, while 4-space indentation is standard in many SQL style guides. Tab indentation lets each developer adjust display width in their own editor.
Why does my subquery output look different from the top-level query?
Subqueries inside parentheses are indented one additional level relative to their surrounding context. This nesting visually communicates the logical hierarchy and makes it easier to trace data flow through complex queries.
Can I format stored procedures or DDL statements?
Yes. The formatter handles CREATE TABLE, ALTER TABLE, INSERT INTO, UPDATE, DELETE, and other DDL/DML statements. CASE expressions, BEGIN/END blocks, and window functions are also indented correctly.
Why are JOIN conditions on a new line?
Placing each JOIN and its ON/USING condition on separate lines makes it clear which conditions belong to which join, especially when a query has four or more joins. This is consistent with the most popular SQL style guides.
How should I handle very long queries?
Paste the full query regardless of length — the formatter works entirely in-browser with no server round-trip, so even large queries with hundreds of lines format instantly. For extremely large queries (> 1 MB) you may notice a brief processing delay.