nakst
[CONTRIB] Init community work to make Luigi the GOTO library for making UIs in **pure C**
I think Luigi has a really great potential to become among the most established library to build (even complex) UIs in pure C. I highlight that as, to me, being able to build UIs without being forced to use the bloated C++ is a HUGE pro.
I mean, C++ is surely handy in speeding up some logic, but I mean, look at projects like UCX, OpenMPI, and I could mention countless I've worked with. No need for C++ if one really commits to it. more effort at the beginning. Triple satisfaction at the end.
Indeed, Luigi still needs a lot of work before even getting closer to what it could be. With this issue I want to open a discussion on this.
Would you consider such idea? It is clear you are not actively maintaining luigi anymore, other than to support the development of GF, as you are carrying modifications to luigi there instead of here. But I guess you are loosing a very big chance to make luigi of much broader reach than it is.
As I was trying to see luigi's fit to build the UI for a library I am working on since some years ton deal with complex Bisepctra computation (and now visualization), I've already started some work, mostly since it could really feel natural to navigate the code when you are not the main creator of it and you it by heart. It is 90% code refactoring, where I separate logically independent concepts into their dedicated TUs. I also isolated the platform specific code (to create windows and rendering mostly, for now) form the UI core itself. This helps eventually to create a common platform interface, and (i) easy adding support for other platforms, (ii) allow the use to use its own platform (can be whatever) as long as the interface contract is respected. BTW, IDK if you tested on WIndows, but I could not compile on WinOS with MSVC. FIxed that.
I have several ideas, but I think we can get to start an open debate once I know I won't be wasting time on this if you are not interested in this.
