kamera 0.0.2

Camera API with a reduced feature set for basic use cases and learning
Documentation

Camera API with a reduced feature set for basic usecases and learning.

  • 🚧 Mac support is based on AVFoundation

  • 🚧 Windows support is based on MediaFoundation

  • 🚧 Linux support is based on V4L2

  • ❌ tests need to run with a single thread cargo t -- --test-threads=1 --nocapture and it is good to review the output of the test cases

  • ✔️ PR check is manual running tests on Mac, Windows and Linux laptop internal camera device and a Logitech external camera

  • ✔️ CI runs checks, formatting and clippy for main and PRs

use kamera::Camera;

let camera = Camera::new_default_device();
camera.start();

let Some(frame) = camera.wait_for_frame() else { return }; // always blockingly waiting for next new frame
let (w, h) = frame.size_u32();

frame.data().data_u32() // use this buffer, per default in ARGB format
// for real use cases processing or displaying frames can get more complicated when trying to be most efficient

camera.stop() // or drop it

Linux system dependecies

On a Debian like system (MX Linux for example) I needed to install these system dependencies to build all crates:

sudo apt install build-essentials cmake libfontconfig1-dev clang
  • TODO winit depends on sctk-adwaita which transitively also is depending on freetype for font rendering. I think
    • winit probably does not have a real need to do font rendering in a default or minimal configuration
    • basic font rendering could also be done by a pure rust dependency