+1 for SQLite! I've used it from Clojure, via HoneySQL, so no ORM, no danger of SQL injection. It was really wonderful! https://github.com/seancorfield/honeysq

+1 for SQLite! I've used it from Clojure, via HoneySQL, so no ORM, no danger of SQL injection. It was really wonderful! https://github.com/seancorfield/honeysql https://github.com/seancorfield/honeysql I used it to quickly iterate on the development of migration SQL scripts for a MySQL DB, which was running in production on RDS. I might have switched to H2 DB later, because that was more compatible with MariaDB, but I could use the same Clojure code, representing the SQL queries, because HoneySQL can emit different syntaxes. Heck, we are even using it to generate queries for the SQL-variant provided by the QuickBooks HTTP API! :) https://www.hugsql.org/ https://www.hugsql.org/ it's pretty good too, btw! it's just a bit too much magic for me personally :) Also, you should really look into JetBrains database tooling, like the one in IntelliJ Ultimate or their standalone DataGrip product! It's freaking amazing, compared to other tools I tried. If you are an Emacs person, then I think even with some inferior shells to the command-line interfaces of the various SQL system, you can go very far a lot more conveniently, than thru some ORMs. Either way, one secret to developing SQL queries comfortably is to utilize some more modern features, like the WITH clause, to provide test data to your queries: https://www.sqlite.org/lang_with.html https://www.sqlite.org/lang_with.html You can use it to just type up some static data, but you can also compute test data dynamically and even randomly! Other little-known feature is the RETURNING clause for INSERT/UPDATE/DELETE: https://www.sqlite.org/lang_returning.html https://www.sqlite.org/lang_returning.html It can highly simplify your host-code, which embeds SQL, because you don't have to introduce UUID keys everywhere, just so you can generate them without coordination.