cosmic VS zig

What if you could develop native applications in JavaScript or WASM, with a runtime that is not Electron? Cosmic is one answer to that, and a project to keep an eye on for sure:

Example can be found at https://github.com/fubark/cosmic/blob/master/ui/examples/text_demo.zig

Definitely want to target Vulkan/Metal/DirectX. Also would like to support WebGPU when it's ready. The drawing is conveniently abstracted away in https://github.com/fubark/cosmic/blob/master/graphics/src/graphics.zig

More details about the UI engine and how to use it in your own zig projects can be found here: https://github.com/fubark/cosmic/tree/master/ui

Yup, I am that matklad. I have thoughts, will put then into a post some time soonish.

zig-analyzer is great — much further along than I’d expected at this point. Though, I am still curious what the end game IDE support would look. https://github.com/ziglang/zig/issues/3028 is a big component of that.

I'm more interested in Zig right now, it looks like a great way to incrementally improve an existing C or C++ code base. I haven't toyed with it yet, but I'd be eager to try it in a project.

correction: zig does not have a built-in, special hash table. Both ArrayHashMap and HashMap are implemented in the standard library.

I'm currently getting into zig and I am trying to make a flake for a test application and i need a zig library mach-glfw for windowing support. Zig is in the process of getting a package manager so for now an idiomatic way of working with 3rd party libraries seems to be to add them as a git submodule in a folder called libs in the project root directory and link them from build.zig

Depends on what you mean by embedded? Microcontrollers (AVG/Arduino)? System-on-a-Chip (SoC)? Embedded Linux system (Raspberry Pi, ARM/MIPS, etc)? There is mrubyc which has been put onto various embedded systems. However, if you want raw performance and minimal memory usage, C is probably still the best option; although it is notoriously difficult to write secure code in C, which is why Rust and Zig are becoming popular for embedded development.

Take a look at the implementation of ArrayList. It allocates a buffer of .capacity sizeof(T) and maintain a slice (.items) that fits exactly the currently used capacity. myList.capacity is the capacity, and myList.items.len is the used-capacity. Not the same name.