diesel VS sqlx

Please reach out on the diesel discussion forum[1] about the lacking dev experience. I'm happy to discuss these issues and potential solutions there.

[1] https://github.com/diesel-rs/diesel/discussions

As is customary for most web frameworks, Rocket allows you to use object-relational mappers (ORMs) as a data access layer for your application. Rocket is ORM agnostic, which means you can use any Rust ORM of your choice to access your database in your Rocket application. In this article, we’ll use Diesel ORM as our ORM of choice as it’s one of the most popular Rust ORMs. At the time of writing, Diesel ORM supports PostgreSQL, MySQL, and SQLite databases.

The Diesel getting started guide is explicitly written for using PostgreSQL as your database system. It is using some parts of SQL that are not supported by SQLite. You can follow the equivalent SQLite code by looking at this diesel example. In the concrete case that error is caused by the fact that only quite new SQLite versions support returning clauses. These support is behind an off-by-default feature flag and requires using an up to date SQLite version.

“watched prime yell about VIM for hours” lmao well done, this got me. A lot of the resources people have already suggested are great! One thing I like to do after going through the documentation for a language is trying to build something like a simple rest api to get a better feel. Rocket (https://rocket.rs) and actix web (https://actix.rs) are popular web frameworks and then for ORM diesel (https://diesel.rs).

You should minimise the non-Rust parts though, don't get tied up in anything else or your scope will quickly outgrow a two month deadline. Use something like https://dummyjson.com/ as a foundation to skip the need to find/generate/build database + content, or use a combination of postgres/postgrest to get a quick API + database and use faker to fill it up quickly. That'd allow you to use something like SQLx or https://diesel.rs/ in the application.

diesel

Congratulations to the release. I know all of this is hard work. I would like to invite you to submit a ormlite implementation to the diesel benchmark collection. As soon as that's merged you will get regular reports here. The relevant code is here in the diesel repository.

Doing arbitrary syntax transformations in procedural macros is not powerful enough? Does C++ have something like serde or diesel?

I've been building a relatively small project with axum and diesel, and I'm pretty happy with it.

There's also Diesel and SeaORM for higher level interfaces.

Perhaps Rust has some nice crates that will assist your development, but you're going to have to evaluate whether their design decisions meet the needs of your design, especially if you want to handle powerloss without data loss whilst maintaining high performance. Here's an example of where a create chose design over performance https://github.com/launchbadge/sqlx/issues/2007

SQLx's query macros act like format macros, and the database is apparently smart enough to not treat bind parameters as part of the SQL.

> Doesn't it need to run introspection queries to check the table structures?

Yes, but it can also run in 'offline' mode where it builds against saved metadata [0]

0: https://github.com/launchbadge/sqlx/blob/main/sqlx-cli/READM...

well it does have its usecases. e.g. https://github.com/launchbadge/sqlx

I felt the same way until I found sqlx (https://github.com/launchbadge/sqlx). The package lets you write sql in your app with compile-time checks. Was so much easier to get up and running than learning a new ORM.

If writing SQL directly, what process do you use to update your queries during schema changes? Do you rely on a test suite to catch errors then update queries by hand? Are you using compile-time checks through libraries like sqlx [1]?

[1]: https://github.com/launchbadge/sqlx

You probably know about this already but we've finally landed the change to one file per query in offline mode which should provide both performance and ergonomic improvements, especially when it comes to merging multiple Git branches that change queries.

What I really like about it (apart from being a really nicely designed language, that is very expressive, powerful, performant and one of the safest because of the strict typing/memory management), is that you can kind of focus on just programming, without all the hassles around setting up a project, thinking about building/deploying etc. as tooling is really awesome as well (rust-analyzer, cargo, crates.io etc.). Libraries are usually high-quality and innovative (which is IMHO not so true for a lot of different other languages, including the ones you mentioned). E.g. if you want to create a web-server/API you could try something like this (my current recommendation): https://github.com/tokio-rs/axum and https://github.com/launchbadge/sqlx for good integration of typed sql in Rust or if you want something higher level: https://github.com/SeaQL/sea-orm

It doesn't allow you to install third-party binaries, but it allows you to make your own binaries runnable with something similar to npx. And, as sqlx-cli is also available as a library, you can make an xtask which just runs that.