There’s a question that always comes up when people pick up the Rust programming language: why are there two string types? Why is there String, and &str?

My Declarative Memory Management article answers the question partially, but there is a lot more to say about it, so let’s run a few experiments and see if we can conjure up a thorough defense of Rust’s approach over, say, C’s.

Background

I recently had a bit of impromptu disaster recovery, and it gave me a hunger for more! More downtime! More kubernetes manifest! More DNS! Ahhhh!

The plan was really simple. I love dedicated Hetzner servers with all my heart but they are not very fungible.

You have to wait entire minutes for a new dedicated server to be provisioned. Sometimes you pay a setup fee, et cetera. And at some point to server static websites and serve as a K3S server, it’s simply just too big, and approximately twice the price that I should pay.

The Nature weekly journal of science was first published in 1869. And after one and a half century, it has finally completed one cycle of carcinization, by publishing an article about the Rust programming language.

It’s a really good article.

What I liked about this article is that it didn’t just talk about performance, or even just memory safety - it also talked about correctness.

I still get excited about programming languages. But these days, it’s not so much because of what they let me do, but rather what they don’t let me do.

Ultimately, what you can with a programming language is seldom limited by the language itself: there’s nothing you can do in C++ that you can’t do in C, given infinite time.

As long as a language is turing-complete and compiles down to assembly, no matter the interface, it’s the same machine you’re talking to. You’re limited by… what your hardware can do, how much memory it has (and how fast it is), what kind of peripherals are plugged into it, and so on.

Now that our development environment is all set up, let’s make something useful!

Creating the catscii crate

From a VS Code window connected to our VM (as we just set up), let’s make a new Rust project:

amos@miles:~$ cargo new catscii Created binary (application) `catscii` package

And open it in a new VSCode window:

amos@miles:~$ code catscii

Even though we’re running this “code” command in the VM guest, VSCode set up enough plumbing that it communicates back to the host, telling it to open another VSCode window, connected to the same SSH remote, in the right folder.

I often write pieces that showcase how well Rust can work for you, and how it can let you build powerful abstractions, and prevents you from making a bunch of mistakes.

If you read something like Some mistakes Rust doesn’t catch in isolation, it could seem as if I had only nice things to say about Rust, and it’s a perfect little fantasy land where nothing ever goes wrong.

It took a hot minute.

Try several weeks.

Well, yeah. I got to contribute to a bunch of open-source projects in the meantime though, so I’m fairly pleased with it!

• libffi (for static linking)
• cairo (more static linking!)
• proxy-libintl (more static linking!)
• expat (static linking strikes again)
• poppler (for file descriptor stuff not properly gated on Windows, closed in favor of a similar MR)

One could say I have a bit of an obsession with build times.

I believe having a “tight feedback loop” is extremely valuable: when I work on a large codebase, I want to be able to make small incremental changes and check very often that things are going as expected.

Especially if I’m working on a project that needs to move quickly: say, the product for an early-stage startup, or a side-project for which I only ever get to do 1-hour work bursts at most.

In the previous article, We’ve built a nagaqueen-based tool that can parse one ooc file, detect class declarations and print its doc strings. Today, we’re making a bit of infrastructure for our app to support more sizable projects.

Source path and lib folders

Parsing a single file was a nice milestone, but it’s not nearly enough. We want to generate documentation for a whole project at a time: and since we’ll want to cross-link the various bits of documentation we generate, we’ll also need to parse the various dependencies (such as the ooc sdk, and any used library) so that we can resolve argument types and link them properly.

I love diagrams. I love them so much!

In fact, I have fairly poor visualization skills, so making a diagram is extremely helpful to me: I’ll have some vague idea of how different things are connected, and then I’ll make a diagram, and suddenly there’s a tangible thing I can look at and talk about.

Of course the diagram only represents a fraction of what I had in mind in the first place, but that’s okay: the point is to be able to talk about some aspect of a concept, and so I have to make choices about what to include in the diagram. And maybe make several diagrams.