easing-fixed: fixed-point easing iterators for rust

This is a fork of https://github.com/joliv/easing which uses fixed-point math instead. no_std compatible.

Features added in fork

• smoothstep
• at(t, start, distance) and at_normalized(t) methods if you don't want iterators:

use easing_fixed::{Fix, ExpInOut};

let res = ExpInOut::at(Fix::from_num(0.5), Fix::from_num(10.), Fix::from_num(100)); // range is 10..=110, result is 60
let res = ExpInOut::at_normalized(Fix::from_num(0.75)); // range is 0..1, result is 0.984375

Development

The test cases define an acceptable error margin (ERROR_MARGIN_FAC) and every test case stepping outside of that threshold gets its "is" and "ought" data written to jupyter-tests/<name of test>-{is,ought}.json. In there, you can use the supplied tests.ipynb notebook for a visualized representation of the failure:

The JSON files get cleared out by build.rs. Sadly on every build, not just during tests, as cfg-gating for test is not possible there.

(mostly preserved) original README below

Rust iterators are fantastic—why not use them for easing? Use like so:

use easing_fixed as easing;

let begin: f64 = 0;
let end:   f64 = 1000;
let steps: u64 = 10;

// In a for loop!
for x in easing::sin_in(begin, end, steps) {
    println!("Woo, I'm at x={}!", x);
}

// Mapping!
fn move_to(x: f64) {
    println!("Moving on up to {}", x);
}
easing::cubic_out(begin, end, steps).map(move_to);

// And other iterator stuff!
let xs: Vec<f64> = easing::exp_inout(begin, end, steps).collect(); // why would you do this

Available iterators are currently:

Easing functions taken from warrenm's AHEasing.