banish

Banish generates a fixed-point runtime at compile time and schedules your states through it. You declare states and rules, the framework decides when they run. Release 1.3.0 adds async support, runtime dispatch, block attributes, transition guards, and scoped variable declarations, rounding out what I think is a genuine framework-level feature set for a category of problem that usually ends up as a loop wrapping a match wrapping a pile of flags. The before/after comparison in the README shows the difference pretty concisely. Happy to answer questions about the design tradeoffs, particularly around the proc macro implementation. Thanks for reading.

I’ve been working on a Rust macro called Banish for defining rule-based state machines declaratively. Instead of manually writing transition loops, Banish evaluates rules within a state until no rule fires (a fixed-point model), then transitions. The macro expands into regular Rust, which allows for seamless integration. The goal is to make complex rule logic and state machines easier to express while keeping runtime costs identical to handwritten code. The project was featured as Crate of the Week in the Rust newsletter this week. I'd love to hear your feedback. Example: use banish::banish; fn main() { let buffer = ["No".to_string(), "hey".to_string()]; let target = "hey".to_string(); let idx = find_index(&buffer, &target); print!("{:?}", idx) } fn find_index(buffer: &[String], target: &str) -> Option<usize> { let mut