Getting Started #

SELECT is an SQL client shaped like an IDE. Schema-aware completion, real-time linting, git-based team workspaces with granular DB permissions, local and proxified connections.

This guide takes you from download to your first query in about five minutes.

The application: the schema tree opened down to a table’s columns, a query mid-completion with the column list over it, its results in the grid below, and the chat answering a question about that same query.

1. Download #

Current release: v0.0.1

See all releases.

Each platform ships as a .zip. The links below always resolve to the latest release; for a specific version swap latest/download for download/vX.Y.Z.

Platform Download
macOS (Apple Silicon) selectDb-darwin-arm64.zip
macOS (Intel) selectDb-darwin-amd64.zip
Windows (x64) selectDb-windows-amd64.zip
Linux (x64) selectDb-linux-amd64.zip

2. Install #

macOS #

Unzip and drag SELECT to your Applications folder.

Windows #

Unzip and run SELECT.exe.

Linux #

Unzip and run the select binary.

3. Open a folder #

A workspace is a folder on your machine, so SELECT starts by asking for one. Pick an empty folder to try it out, or a repository of .sql files you already have. SELECT adds a single select.config.json naming the workspace and leaves everything else alone; an empty folder also gets a small sample database to run queries against.

See Workspace Folder for sharing one with a team.

4. Connect your first database #

SELECT supports PostgreSQL, MySQL, and SQLite.

  1. Right-click a folder in the filesystem panel and choose New Database….

  2. Pick your dialect and fill in the DSN:

    Dialect DSN format
    PostgreSQL host=localhost port=5432 user=admin password=*** dbname=mydb sslmode=require
    MySQL user:password@tcp(host:3306)/dbname?parseTime=true
    SQLite file:./local.db or /path/to/database.sqlite

  3. Click Test connection to check it. There is nothing to save: the form writes datasource.config.json for you once the connection is valid.

The connection form: dialect, name, and a DSN written as $WAREHOUSE_DSN rather than a literal connection string.

DSN fields support environment variables ($VAR_NAME), resolved from your workspace .env file at connection time. Keep credentials in .env so you can safely commit your connection config to git.

For full details, see Connecting a Database.

5. Run your first query #

  1. Right-click your database folder and choose New file… to create a .sql file. (In a hurry? Cmd+N opens a scratch tab you can save later.)

  2. Press Cmd+Shift+D to open the database picker and associate the file with the connection you just created.

    The database picker open over a SQL file, with the warehouse connection ticked.

  3. Start typing. Completion comes from the schema SELECT just read off your database, so after FROM you get your own tables, and after a table you get its columns with their types. Misspell one and the linter marks it before you run anything:

    SELECT * FROM your_table LIMIT 10;

  4. Press Cmd+Enter to run. Results appear in a table below the editor.

    A query run: twelve rows of weekly revenue in the result grid, with the time it took beside the row count.

You’re up and running.

Where to go next

Edit this page on GitHub