frewsxcv / aho-corasick Goto Github PK

This crate provides an implementation of the Aho-Corasick algorithm. Its intended use case is for fast substring matching, particularly when matching multiple substrings in a search text. This is achieved by compiling the substrings into a finite state machine.

This implementation provides optimal algorithmic time complexity. Construction of the finite state machine is O(p) where p is the length of the substrings concatenated. Matching against search text is O(n + p + m), where n is the length of the search text and m is the number of matches.

Dual-licensed under MIT or the UNLICENSE.

http://burntsushi.net/rustdoc/aho_corasick/.

The documentation contains several examples, and there is a more complete example as a full program in examples/dict-search.rs.

Here is a quick example showing simple substring matching:

use aho_corasick::{Automaton, AcAutomaton, Match};

let aut = AcAutomaton::new(vec!["apple", "maple"]);
let mut it = aut.find("I like maple apples.");
assert_eq!(it.next(), Some(Match {
    pati: 1,
    start: 7,
    end: 12,
}));
assert_eq!(it.next(), Some(Match {
    pati: 0,
    start: 13,
    end: 18,
}));
assert_eq!(it.next(), None);

Aho-Corasick is useful for matching multiple substrings against many long strings. If your long string is fixed, then you might consider building a suffix array of the search text (which takes O(n) time). Matches can then be found in O(plogn) time.