You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 
gitea/vendor/github.com/blevesearch/vellum/levenshtein
techknowlogick 47f6a4ec3f
go1.16 (#14783)
4 years ago
..
LICENSE go1.16 (#14783) 4 years ago
README.md go1.16 (#14783) 4 years ago
alphabet.go go1.16 (#14783) 4 years ago
dfa.go go1.16 (#14783) 4 years ago
levenshtein.go go1.16 (#14783) 4 years ago
levenshtein_nfa.go go1.16 (#14783) 4 years ago
parametric_dfa.go go1.16 (#14783) 4 years ago

README.md

levenshtein

levenshtein automaton

This package makes it fast and simple to build a finite determinic automaton that computes the levenshtein distance from a given string.

Sample usage:

// build a re-usable builder
lb := NewLevenshteinAutomatonBuilder(2, false)

origTerm := "couchbasefts"
dfa := lb.BuildDfa("couchbases", 2)
ed := dfa.eval([]byte(origTerm))
if ed.distance() != 2 {
	log.Errorf("expected distance 2, actual: %d", ed.distance())
}

This implementation is inspired by blog post and is intended to be a port of original rust implementation: https://github.com/tantivy-search/levenshtein-automata

Micro Benchmark Results against the current vellum/levenshtein is as below.

BenchmarkNewEditDistance1-8       	   30000	     52684 ns/op	   89985 B/op	     295 allocs/op
BenchmarkOlderEditDistance1-8     	   10000	    132931 ns/op	  588892 B/op	     363 allocs/op

BenchmarkNewEditDistance2-8       	   10000	    199127 ns/op	  377532 B/op	    1019 allocs/op
BenchmarkOlderEditDistance2-8     	    2000	    988109 ns/op	 4236609 B/op	    1898 allocs/op