2016-08-23 23:06:30 -04:00
2016-08-23 21:34:12 -04:00
2016-08-23 22:53:08 -04:00
2015-06-10 00:26:44 -04:00
2016-06-05 03:47:56 -04:00
2016-06-05 03:47:56 -04:00
2015-10-01 00:01:05 -04:00
2016-02-20 12:31:03 -05:00
2016-08-11 21:43:06 -04:00
2016-08-23 23:06:30 -04:00
2015-02-09 20:04:10 -05:00
2016-08-23 23:06:30 -04:00
2016-01-29 23:46:17 -05:00

strsim-rs Crates.io Crates.io Linux build status Windows build status

Rust implementations of string similarity metrics:

Installation

# Cargo.toml
[dependencies]
strsim = "0.5.1"

Documentation

Usage

extern crate strsim;

use strsim::{hamming, levenshtein, damerau_levenshtein, jaro, jaro_winkler,
             levenshtein_against_vec, damerau_levenshtein_against_vec,
             jaro_against_vec, jaro_winkler_against_vec};

fn main() {
    match hamming("hamming", "hammers") {
        Ok(distance) => assert_eq!(3, distance),
        Err(why) => panic!("{:?}", why)
    }

    assert_eq!(3, levenshtein("kitten", "sitting"));

    assert_eq!(1, damerau_levenshtein("specter", "spectre"));

    assert!((0.392 - jaro("Friedrich Nietzsche", "Jean-Paul Sartre")).abs() <
            0.001);

    assert!((0.911 - jaro_winkler("cheeseburger", "cheese fries")).abs() <
            0.001);

    // get vectors of values back
    let v = vec!["test", "test1", "test12", "test123", "", "tset"];

    assert_eq!(levenshtein_against_vec("test", &v),
               vec![0, 1, 2, 3, 4, 2]);

    assert_eq!(damerau_levenshtein_against_vec("test", &v),
               vec![0, 1, 2, 3, 4, 1]);

    let jaro_distances = jaro_against_vec("test", &v);
    let jaro_expected = vec![1.0, 0.933333, 0.888889, 0.857143, 0.0, 0.916667];
    let jaro_delta: f64 = jaro_distances.iter()
                                        .zip(jaro_expected.iter())
                                        .map(|(x, y)| (x - y).abs() as f64)
                                        .fold(0.0, |x, y| x + y as f64);
    assert!(jaro_delta < 0.0001);

    let jaro_winkler_distances = jaro_winkler_against_vec("test", &v);
    let jaro_winkler_expected = vec![1.0, 0.96, 0.933333, 0.914286, 0.0, 0.925];
    let jaro_winkler_delta = jaro_winkler_distances.iter()
                                 .zip(jaro_winkler_expected.iter())
                                 .map(|(x, y)| (x - y).abs() as f64)
                                 .fold(0.0, |x, y| x + y as f64);
    assert!(jaro_winkler_delta < 0.0001);
}

Development

Install Vagrant, and run vagrant up.

License

MIT

Description
No description provided
Readme MIT 738 KiB
Languages
Rust 96.7%
Python 3.3%