using slint in a Rust ESP-IDF component

[floers]

Hey,

I am currently trying to use slint with epdiy. Since I don't want to use C I created a project as described here: https://github.com/esp-rs/esp-idf-template/blob/master/README-cmake.md and then I added slint to the inner rust project now I can't compile anymore. I get lots of errors that transitive dependencies could not be built. Like this:

error[E0433]: failed to resolve: could not find `Float` in `num_traits`
  --> /home/florian/.cargo/registry/src/index.crates.io-1949cf8c6b5b557f/euclid-0.22.11/src/trig.rs:48:41
   |
48 |                 let y_abs = num_traits::Float::abs(y);
   |                                         ^^^^^ could not find `Float` in `num_traits`
...
80 | trig!(f64);
   | ---------- in this macro invocation
   |

I added the slint dependencies as described here:

https://docs.slint.dev/latest/docs/slint/guide/platforms/embedded/

But I am not sure whether this can work ATM. Can I even apply the #[mcu_board_support::entry] in this setup?

I wanted to use the software renderer to write into a simple buffer (similar to this but no_std). Everything from threre on already works.

My current status: https://gitlab.com/loers/epdiy-rs-example

Has anyone ideas what I am missing or if this is even possible currently?

[tronical]

Hmm, one thing that I see missing is that you’re not enabling any features in the Slint crate. You might need at least „compat-1-2“ and „software-renderer“, possibly more.

[floers]

The issue was that the Float trait in num-traits (transitive dependency from slint) is only enabled on features std and libm. Which I guess aren't active when building slint without slints std feature...

So I added this to my Carg.toml:

slint = { version = "1.13.1", default-features = false, features = ["compat-1-2", "renderer-software", "unsafe-single-threaded"] }
[dependencies.num-traits]
version = "0.2"
features = ["libm"]

And now I can at least compile.

[tronical]

Right, that makes sense. The slint crate also has a „libm“ feature for this purpose. From the docs:

This feature enables floating point arithmetic emulation using the libm crate. Use this
in MCU environments where the processor does not support floating point arithmetic.
libm = […]