taffy

taffy is a flexible, high-performance, cross-platform UI layout library written in Rust.

Currently Taffy implements only CSS based layout algorithms:

• The stable 0.2.x releases of Taffy implements the flexbox layout algorithm.
• Support for CSS Grid is in preview. If you wish to try this out then you should use the 0.3.x alpha releases and enable the experimental_grid cargo feature (from 0.3.0-alpha2 the CSS Grid feature is enabled by default). For information, see the release notes and the tracking issue. Experimentation with the CSS Grid implementation is encouraged, and feedback and bug reports are welcomed.

Support for other paradigms is planned.

This crate is a collaborative, cross-team project, and is designed to be used as a dependency for other UI and GUI libraries. Right now, it powers:

• Dioxus: a React-like library for building fast, portable, and beautiful user interfaces with Rust
• Bevy: an ergonomic, ECS-first Rust game engine

Usage

use taffy::prelude::*;

// First create an instance of Taffy
let mut taffy = Taffy::new();

// Create a tree of nodes using `taffy.new_leaf` and `taffy.new_with_children`.
// These functions both return a node id which can be used to refer to that node
// The Style struct is used to specify styling information
let header_node = taffy
    .new_leaf(
        Style {
            size: Size { width: points(800.0), height: points(100.0) },
            ..Default::default()
        },
    ).unwrap();

let body_node = taffy
    .new_leaf(
        Style {
            size: Size { width: points(800.0), height: auto() },
            flex_grow: 1.0,
            ..Default::default()
        },
    ).unwrap();

let root_node = taffy
    .new_with_children(
        Style {
            flex_direction: FlexDirection::Column,
            size: Size { width: points(800.0), height: points(600.0) },
            ..Default::default()
        },
        &[header_node, body_node],
    )
    .unwrap();

// Call compute_layout on the root of your tree to run the layout algorithm
taffy.compute_layout(root_node, Size::MAX_CONTENT).unwrap();

// Inspect the computed layout using taffy.layout
assert_eq!(taffy.layout(root_node).unwrap().size.width, 800.0);
assert_eq!(taffy.layout(root_node).unwrap().size.height, 600.0);
assert_eq!(taffy.layout(header_node).unwrap().size.width, 800.0);
assert_eq!(taffy.layout(header_node).unwrap().size.height, 100.0);
assert_eq!(taffy.layout(body_node).unwrap().size.width, 800.0);
assert_eq!(taffy.layout(body_node).unwrap().size.height, 500.0); // This value was not set explicitly, but was computed by Taffy

Benchmarks (vs. Yoga)

• Run on a 2021 MacBook Pro with M1 Pro processor.
• Taffy benchmarks are using criterion (10 iterations).
• Yoga benchmarks run via it's node.js bindings (the yoga-layout-prebuilt npm package), they were run a few times manually and it was verified that variance in the numbers of each run was minimal. It should be noted that this is using an old version of Yoga.

(note that the table below contains multiple different units (milliseconds vs. microseconds vs. nanoseconds))

Benchmark	Yoga	Taffy 0.2
yoga/10 nodes (1-level hierarchy)	45.1670 µs	1.9857 µs
yoga/100 nodes (2-level hierarchy)	134.1250 µs	41.810 µs
yoga/1_000 nodes (3-level hierarchy)	1.2221 ms	357.48 µs
yoga/10_000 nodes (4-level hierarchy)	13.8672 ms	3.7310 ms
yoga/100_000 nodes (5-level hierarchy)	141.5307 ms	39.682 ms

Most popular websites seem to have between 3,000 and 10,000 nodes (although they also require text layout, which neither yoga nor taffy implement).

Contributions

Contributions welcome: if you'd like to use, improve or build taffy, feel free to join the conversation, open an issue or submit a PR. If you have questions about how to use taffy, open a discussion so we can answer your questions in a way that others can find.