nix VS cxx

If we drop std Rust ceases to be economical due to the time it would take to reimplement the data structures and IO interfaces it provides, not to mention the event loop crate we use (calloop). At that point we'd be relying on so much FFI via eg. nix that the relative safety would be diminished too. After reimplementing all that it's not clear to me that we'd even save that much size, but I suppose it's possible.

Now that we have covered the fundamentals of signals, let's delve into the world of handling signals in Rust! Unlike C, where signal handling is built into the language modules, Rust provides several libraries that enable developers to handle signals with ease. Libraries such as signal_hook, nix, libc, and tokio handle signals that primarily use C bindings to make it possible to work with signals.

Obviously you meant the nix crate

Finally, there's the nix crate, which provides a safe Rust API over the libc functions.

You can take a look here for example: https://github.com/nix-rust/nix/blob/master/CHANGELOG.md

Unsafe code can in principle speed up I/O by calling libc for special syscalls, but uutils typically uses safe wrappers from nix instead. Very rarely there's a line of unsafe code needed to sand off the edges.

This post fails to speak to me on two fronts:

* The `nix` crate is a cornerstone of the Rust development ecosystem: if you do anything that requires POSIX or various nix-specific APIs beyond those wrapped by the standard library, then `nix` most likely provides a high-level and safe* wrapper for them. Perfectionism is a virtue in this context, one that keeps large parts of the Rust ecosystem from accidentally consuming buggy code. The author unfortunately chose a particularly messy and bug-prone corner of the POSIX APIs to wrap, and ran into a correspondingly intensive review process. I've merged simpler wrappers[1][2] with no fuss.

* Alacritty seems to work just fine. I switched to it about two months ago, after using nothing but (heavily customized) rxvt-unicode for a decade. Maybe it's because I don't use ligatures or images in my terminals (I thought we were talking about non-"toy" functionality!), but I haven't found myself wanting for anything beyond what Alacritty already does. And the scrollback seems to work nicely. To summarize: where's the tragedy?

[1]: https://github.com/nix-rust/nix/pull/1342

[2]: https://github.com/nix-rust/nix/pull/1331

One thing I'd like to see is a MSRV policy, as its causing problems for downstreams (https://github.com/nix-rust/nix/issues/1555)

There are cxx and autocxx, what else do you propose to do?

I would like to see a comparison of how this compares to Rust. In terms of interoperability it has Cxx (https://cxx.rs) to offer safe bindings to C++ but also has great support for Android, Linux and many other systems. You don't even need to hack together Windows bindings (as explained in the blog post) because Microsoft offers official bindings (https://crates.io/crates/windows). I'm not sure if I'd call it a superpower if any potential interoperability has to be written to be used (compared to it already being available). Or rather, in comparison to what is interoperability a Swift superpower? Certainly not C++ or C which can be used in a far wider set of targets.

We selected Qt as a cross-platform solution. The C++/Rust interface is the clunkiest and ugliest part of the application, and rather complex because some state is shared between several windows in the GUI and several threads in the backend, and any component might modify that state at any time, and updates have to be transmitted to the other components without introducing inconsistencies. Using cxx [1] helped a little, though.

The project began in 2020, and I'm not sure what I'd choose as a GUI framework today – definitely not Qt Widgets, though.

[1] https://cxx.rs/

If the build process for the C library isn't too involved I recommend using cxx bridge (https://cxx.rs/) and letting cargo handle the build and linking. cxx basically allows you to describe the bidirectional interface (although it sounds like you only need 1 direction, which is fine too) in Rust code and it provides a "good enough" API for compiling C code inside the build.rs file.

The tooling for the first kind -- calling Rust from another language -- is a bit less developed, and tends to rely on code generation that doesn't necessarily produce a natural C API. cbindgen, uniffi, cxx, and Diplomat all take this course.

I would like to utilize OMPL's functionality in Rust code, so I want to call into OMPL C++ code somehow in Rust. I've seen two (non-mutually-exclusive) options so far: - rust-cpp, which allows you to write C++ code in Rust within the cpp!() macro. - cxx, which allows you to define both sides of the FFI boundary manually (as opposed to bindgen's automatic generation).

I'm not sure how to do this in cxx; issues like https://github.com/dtolnay/cxx/issues/447 suggest that this isn't settled yet?

I use non-vendored dependencies for the Buck build in https://github.com/dtolnay/cxx.

There's also the cpp and cxx crates for doing C++/Rust interop, but they probably aren't appropriate to use in all cases. The C ABI is definitely the safest way to go unless you're really trying to marry Rust and C++ code bases, not just writing library bindings.

There's also cxx (can't vouch for it personally but it claims to make things a lot easier) https://github.com/dtolnay/cxx