umka-lang VS jstar

I'm a fan or Umka. It is inspired by Go and can be embedded into C programs.

https://github.com/vtereshkov/umka-lang

My scripting language Umka now supports closures. This is not merely a matter of fashion, but the requirement of the Tophat game engine that heavily relies on callbacks, particularly for inter-module communication via signals.

After three years of development, I released Umka 1.0, a statically typed scripting language designed for embedding into C/C++ host applications. Its syntax and some key features were inspired by Go. However, Umka doesn't rely on the Go ecosystem and only needs the C standard library to run.

Yaegi needs to be embedded in another golang program. It’s a golang module to evaluate golang. Which is cool, but different from what Umka seems to be.

Umka is a scripting language with syntax inspired by golang. A language with its own vm that you can embed anywhere, exsmple: https://github.com/vtereshkov/umka-lang/blob/master/examples....

Yaegi and Umka are two different things.

UmPlot 0.1 has been released. It is an easy-to-use cross-platform plotting library for the Umka language. It is based on raylib and distributed as an Umka Module Implementation (UMI) file, i. e. a language extension shared library with an Umka interface.

Some people who tried to use the Umka scripting language were frustrated by the absence of closures. This feature is under development and not yet released. However, you can easily simulate closures with the good old structures, methods and interfaces. This simple truth sometimes comes as a surprise to the users.

I have been developing the statically typed scripting language Umka for more than a year. Now it is stable enough to serve as the basis for various projects: a 2D game engine, a VDrift-based framework for exercising with the car autopilot logic, and a proprietary tractor dynamics simulator.

[LAUNGUAGE: J*]

My language, obviously.

Also, since you are willing to look at non-treewalk interpreters, you can take a look at this: https://github.com/bamless/jstar/blob/master/src/vm.h In this I use two parallel arrays, one for frames and one for variables. The frame contains the instruction pointer of the function and the address of the start of the activation record. Variables are stored packed inside the stack array and are accessed through indexing.