Nifty / Extensions / Data & formats
Nifty SQL
Format SQL in 20 dialects, including dbt and Jinja templates, with your team's settings.
code --install-extension nifty.nifty-sql
Works in VS Code, Cursor, Windsurf, VSCodium and in the browser on vscode.dev and github.dev. Free.
Try it without installing anything: SQL Formatter · PostgreSQL Formatter · MySQL Formatter
Features




- 20 dialects. Standard SQL, PostgreSQL, MySQL, MariaDB, SQL Server (T-SQL), Oracle PL/SQL, SQLite, Snowflake, BigQuery, Redshift, DuckDB, ClickHouse, Spark, Hive, Trino, Db2, Db2 for i, N1QL, SingleStoreDB and TiDB.
- Just use Format Document. Nifty SQL plugs into VS Code's built-in Format Document, Format Selection and format on save for SQL files.
- Format SQL anywhere. Select a query inside Python, TypeScript, Go or any other file and run Nifty SQL: Format SQL. The result keeps the indentation of the surrounding code, and query parameters such as
%(name)s,${id}and:namestay as they are. - dbt and Jinja friendly.
{{ ref('orders') }},{% if is_incremental() %}and{# comments #}survive formatting untouched, and block tags stay on their own lines. - One style for the whole team. Commit a
.sql-formatter.jsonto your repo and everyone gets the same output. It is the same file the sql-formatter CLI reads, so CI can check formatting too. You get autocomplete and validation while editing it. - The right dialect automatically. The dialect comes from the file's language (e.g. files marked MySQL or PostgreSQL by your database extension), and is shown in the status bar (a
languagein.sql-formatter.jsonwins). Click it to set a dialect for the workspace. - Errors you can read. If the SQL can't be parsed you get one line saying where, not a wall of text.
- Works in the browser on vscode.dev and github.dev.
Commands
| Command | What it does |
|---|---|
Nifty SQL: Format SQL (Document or Selection) |
Format the selection, or the whole file if nothing is selected. Works in any language. |
Nifty SQL: Format SQL with Dialect… |
Pick a dialect for a one-off format |
Nifty SQL: Set SQL Dialect for Workspace… |
Save the dialect for this workspace (also available by clicking the status bar) |
Nifty SQL is the default formatter for .sql files. To format on save, add this to your settings:
"[sql]": {
"editor.formatOnSave": true
}
Team configuration: .sql-formatter.json
{
"language": "postgresql",
"keywordCase": "upper",
"tabWidth": 4,
"linesBetweenQueries": 2
}
Nifty SQL uses the nearest .sql-formatter.json above the file (within the workspace). Its values override your personal settings.
Settings
| Setting | Default | What it does |
|---|---|---|
nifty.sql.dialect |
auto |
Dialect, or auto to use the file's language |
nifty.sql.keywordCase |
preserve |
preserve, upper or lower for keywords |
nifty.sql.dataTypeCase |
preserve |
Case for data types |
nifty.sql.functionCase |
preserve |
Case for function names |
nifty.sql.identifierCase |
preserve |
Case for table and column names |
nifty.sql.indentStyle |
standard |
standard, tabularLeft or tabularRight |
nifty.sql.logicalOperatorNewline |
before |
Line break before or after AND / OR |
nifty.sql.expressionWidth |
50 |
Width at which parenthesised expressions are split |
nifty.sql.linesBetweenQueries |
1 |
Empty lines between statements |
nifty.sql.denseOperators |
false |
a+b instead of a + b |
nifty.sql.newlineBeforeSemicolon |
false |
Put the closing ; on its own line |
nifty.sql.templating |
auto |
Jinja/dbt handling: auto, jinja or none |
nifty.sql.useConfigFile |
true |
Read .sql-formatter.json |
nifty.sql.showDialectInStatusBar |
true |
Show the dialect in the status bar |
Indentation (spaces or tabs, and width) follows your editor settings unless .sql-formatter.json sets it.