7 releases

✓ Uses Rust 2018 edition

0.2.3 Oct 14, 2019
0.2.2 Dec 13, 2018
0.2.1 Oct 29, 2018
0.2.0 Aug 31, 2018
0.1.2 Aug 16, 2018

#163 in Data structures

49 downloads per month

MIT license



Build Status Coverage Status MIT License Crate API Docs

Finger trees is a functional representation of persistent sequences supporting access to the ends in amortized constant time, and concatenation and splitting in time logarithmic in the size of the smaller piece. It also has split operation defined in general form, which can be used to implement sequence, priority queue, search tree, priority search queue and more datastructures.



  • This implementation does not use non-regular recursive types as implementation described in the paper. As rust's monomorphization does not play well with such types.
  • Implmentation abstracts over reference counted types Rc/Arc. Using type family trick.
  • Uses strict spine in implementation.
  • Iterator returns cloned value, and in general this implementation assumes that value stored in a tree is cheaply clonable, if it is not you can always put it in a Rc/Arc or anything else.


use std::iter::FromIterator;
use fingertrees::measure::Size;
use fingertrees::monoid::Sum;
use fingertrees::{FingerTree, Measured, RcRefs};

// construct `Rc` based finger tree with `Size` measure
let ft: FingerTree<RcRefs, _> = vec!["one", "two", "three", "four", "five"]
assert_eq!(ft.measure(), Sum(5));

// split with predicate
let (left, right) = ft.split(|measure| *measure > Sum(2));
assert_eq!(left.measure(), Sum(2));
assert_eq!(Vec::from_iter(&left), vec![Size("one"), Size("two")]);
assert_eq!(right.measure(), Sum(3));
assert_eq!(Vec::from_iter(&right), vec![Size("three"), Size("four"), Size("five")]);

// concatinate
assert_eq!(ft, left + right);

// push values
    vec!["left", "one", "two", "three", "four", "five", "right"]

No runtime deps