Tutorials

Design a project-management schema with tasks, labels, and assignments. Covers SERIAL, foreign keys, CHECK constraints, ON CONFLICT upserts, and RETURNING.

Build a sales pipeline and slice it with aggregation, window functions, ROLLUP/CUBE, and reusable views. Rank reps, compute running totals, and generate subtotal rows.

Model sensor readings with timestamps and intervals. Use DATE_TRUNC, EXTRACT, AGE, TO_CHAR, recursive CTEs to fill date gaps, and window frames for moving averages.

An e-commerce schema with users, orders, items, and products. Walk through INNER, LEFT, RIGHT, FULL, CROSS, and NATURAL joins, plus correlated subqueries and EXISTS.

Start with a v1 schema and evolve it: add columns, rename fields, change types with CAST/::, and wrap migrations in transactions with nested BEGIN/COMMIT.

Model an org chart with self-referencing foreign keys and query it with WITH RECURSIVE. Walk trees top-down and bottom-up, track depth, aggregate subtrees, and limit recursion depth.

Compare two warehouse inventories with UNION, UNION ALL, INTERSECT, and EXCEPT. Find shared products, exclusive items, and aggregate combined stock.

Transform a product catalog with UPPER, SUBSTRING, REPLACE, CONCAT, ABS, ROUND, COALESCE, NULLIF, and CASE WHEN.

Create B-tree indexes, use EXPLAIN to see query plans, understand when indexes help and when they hurt. Covers IF NOT EXISTS, primary key indexes, and DROP INDEX.

Peek behind SERIAL to see sequences, create custom sequences with START/INCREMENT, and define ENUM types for fixed-set columns with ordinal ordering.

Map Parquet files to SQL tables with CREATE FOREIGN TABLE. Run joins, aggregates, subqueries, and CTEs across Parquet data. 9 benchmarked workloads on 50K–200K row files. Requires native server (not WASM).