levenshtein: wrap to sha256 for state hashing

go.mod

@@ -4,7 +4,7 @@ go 1.21
 
 require (
 	github.com/bits-and-blooms/bitset v1.2.0
-	github.com/blevesearch/mmap-go v1.0.4
+	github.com/edsrzf/mmap-go v1.2.1-0.20241212181136-fad1cd13edbd
 	github.com/spf13/cobra v1.4.0
 )
 

go.sum

@@ -1,8 +1,8 @@
 github.com/bits-and-blooms/bitset v1.2.0 h1:Kn4yilvwNtMACtf1eYDlG8H77R07mZSPbMjLyS07ChA=
 github.com/bits-and-blooms/bitset v1.2.0/go.mod h1:gIdJ4wp64HaoK2YrL1Q5/N7Y16edYb8uY+O0FJTyyDA=
-github.com/blevesearch/mmap-go v1.0.4 h1:OVhDhT5B/M1HNPpYPBKIEJaD0F3Si+CrEKULGCDPWmc=
-github.com/blevesearch/mmap-go v1.0.4/go.mod h1:EWmEAOmdAS9z/pi/+Toxu99DnsbhG1TIxUoRmJw/pSs=
 github.com/cpuguy83/go-md2man/v2 v2.0.1/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
+github.com/edsrzf/mmap-go v1.2.1-0.20241212181136-fad1cd13edbd h1:I4PrRZuNMeDP3VbFrak4QsqwO5tWkQf0tqrrr1L2DsU=
+github.com/edsrzf/mmap-go v1.2.1-0.20241212181136-fad1cd13edbd/go.mod h1:19H/e8pUPLicwkyNgOykDXkJ9F0MHE+Z52B8EIth78Q=
 github.com/inconshreveable/mousetrap v1.0.0 h1:Z8tu5sraLXCXIcARxBp/8cbvlwVa7Z1NHg9XEKhtSvM=
 github.com/inconshreveable/mousetrap v1.0.0/go.mod h1:PxqpIevigyE2G7u3NXJIT2ANytuPF1OarO4DADm73n8=
 github.com/russross/blackfriday/v2 v2.1.0/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=

levenshtein/parametric_dfa.go

@@ -15,7 +15,7 @@
 package levenshtein
 
 import (
-	"crypto/md5"
+	"crypto/sha256"
 	"encoding/json"
 	"fmt"
 	"math"
@@ -311,13 +311,13 @@ func fromNfa(nfa *LevenshteinNFA) (*ParametricDFA, error) {
 }
 
 type hash struct {
-	index map[[16]byte]int
+	index map[[32]byte]int
 	items []MultiState
 }
 
 func newHash() *hash {
 	return &hash{
-		index: make(map[[16]byte]int, 100),
+		index: make(map[[32]byte]int, 100),
 		items: make([]MultiState, 0, 100),
 	}
 }
@@ -326,9 +326,9 @@ func (h *hash) getOrAllocate(m MultiState) int {
 	size := len(h.items)
 	var exists bool
 	var pos int
-	md5 := getHash(&m)
-	if pos, exists = h.index[md5]; !exists {
-		h.index[md5] = size
+	sha256Hash := getHash(&m)
+	if pos, exists = h.index[sha256Hash]; !exists {
+		h.index[sha256Hash] = size
 		pos = size
 		h.items = append(h.items, m)
 	}
@@ -339,11 +339,11 @@ func (h *hash) getFromID(id int) *MultiState {
 	return &h.items[id]
 }
 
-func getHash(ms *MultiState) [16]byte {
+func getHash(ms *MultiState) [32]byte {
 	msBytes := []byte{}
 	for _, state := range ms.states {
 		jsonBytes, _ := json.Marshal(&state)
 		msBytes = append(msBytes, jsonBytes...)
 	}
-	return md5.Sum(msBytes)
+	return sha256.Sum256(msBytes)
 }

vellum_mmap.go

@@ -12,14 +12,15 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+//go:build !nommap
 // +build !nommap
 
 package vellum
 
 import (
 	"os"
 
-	mmap "github.com/blevesearch/mmap-go"
+	mmap "github.com/edsrzf/mmap-go"
 )
 
 type mmapWrapper struct {