Query from Lakebase SQL Editor

The Lakebase SQL Editor runs queries on your Lakebase databases directly from the Lakebase App. It offers Postgres-native features such as EXPLAIN/ANALYZE, psql-style meta-commands, and exporting results to CSV/JSON/XLSX.

Use the Lakebase SQL Editor

To use the SQL Editor:

1. Open the Lakebase App and select your project.
2. Select SQL Editor from the sidebar, then select a branch and database.
3. Enter a query into the editor and click Run to view the results.

Use the following query to try the SQL Editor. The query creates a table, adds data, and retrieves the data from the table.

CREATE TABLE IF NOT EXISTS playing_with_lakebase(id SERIAL PRIMARY KEY, name TEXT NOT NULL, value REAL);
INSERT INTO playing_with_lakebase(name, value)
SELECT LEFT(md5(i::TEXT), 10), random() FROM generate_series(1, 10) s(i);
SELECT * FROM playing_with_lakebase;

Running multiple query statements at once returns a separate result set for each statement. The result sets are displayed in separate tabs, numbered in order of execution.

To clear the editor, select the contents of the text box and delete it.

Explain and analyze

The Lakebase SQL Editor has Explain and Analyze features.

• The Explain feature runs the specified query with the Postgres EXPLAIN command, which returns the execution plan for the query. The Explain feature only returns a plan with estimates. It does not run the query.
• The Analyze feature runs the specified query with EXPLAIN ANALYZE. The ANALYZE parameter executes the query and returns actual row counts and run times for plan nodes along with the EXPLAIN estimates.

Understanding the information provided by the Explain and Analyze features requires familiarity with the Postgres EXPLAIN command and its ANALYZE parameter. See the EXPLAIN documentation and Using EXPLAIN in the PostgreSQL documentation.

Export data to CSV, JSON, and XLSX

The Lakebase SQL Editor supports exporting your data to JSON, CSV, and XLSX. Access the download button from the lower right corner of the SQL Editor page. The download button only appears when there is a result set to download.

Expand results section of the SQL editor window

Expand the results section of the SQL Editor window by selecting the expand window button from the lower right corner of the SQL Editor page.

Meta-commands

The Lakebase SQL Editor supports using Postgres meta-commands, which act like shortcuts for interacting with your database. If you are already familiar with using meta-commands from the psql command-line interface, you can use many of those same commands in the Lakebase SQL Editor.

Meta-commands can speed up your workflow by providing quick access to database schemas and other critical information without needing to write full SQL queries.

Here are some commonly-used meta-commands within the Lakebase SQL Editor:

• \dt — List all tables in the current database
• \d [table_name] — Describe a table's structure
• \l — List all databases
• \? — A cheat sheet of available meta-commands
• \h [NAME] — Get help for any Postgres command (e.g., \h SELECT)

For a complete list of meta-commands and their usage, see Meta-commands in psql.

How to use meta-commands

To use a meta-command in the SQL Editor, enter the meta-command in the editor (just like a SQL query) and click Run. The result will be displayed in the output pane.

SQL Editor limitations when public access is disabled

When your workspace has Private Link configured with public access disabled, the SQL Editor proxies queries through the Lakebase backend instead of using a direct database connection. Most queries work identically to a direct connection, but the proxy is stateless and runs each statement as an independent HTTP request, which introduces the following limitations:

What works:

• Single-statement queries (SELECT, INSERT, UPDATE, DELETE, UPSERT)
• DDL statements (CREATE, ALTER, DROP TABLE, CREATE INDEX, and others)
• EXPLAIN and DESCRIBE commands (\d, \dt, \di)
• Full Postgres data type support (arrays, JSON/JSONB, numeric, timestamps, booleans)
• Query results and error messages

What does not work:

For workloads that require transactions or session state, connect directly to your database using a Postgres client. See Connect with psql.

Troubleshoot

If the SQL Editor displays a "Failed to fetch" or "Unknown error" message and your workspace uses Private Link, confirm that inbound Private Link (workspace-level, port 443) is correctly configured. See Troubleshoot Private Link connectivity.