New similarity module to help with comparing track titles

internal/similarity/similarity.go

@@ -0,0 +1,55 @@
+/*
+Copyright © 2024 Philipp Wolfer <phw@uploadedlobster.com>
+
+Scotty is free software: you can redistribute it and/or modify it under the
+terms of the GNU General Public License as published by the Free Software
+Foundation, either version 3 of the License, or (at your option) any later version.
+
+Scotty is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
+A PARTICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+Scotty. If not, see <https://www.gnu.org/licenses/>.
+*/
+
+package similarity
+
+import (
+	"regexp"
+	"strings"
+
+	"github.com/agnivade/levenshtein"
+	"go.uploadedlobster.com/scotty/internal/util"
+	"golang.org/x/text/unicode/norm"
+)
+
+// Returns the Levensthein distance between s1 and s2 relative to the length of
+// the longer string.
+// Unicode normalization on the strings is performed.
+func Similarity(s1 string, s2 string) float64 {
+	s1 = norm.NFKC.String(s1)
+	s2 = norm.NFKC.String(s2)
+	l1 := len([]rune(s1))
+	l2 := len([]rune(s2))
+	maxLen := util.Max(l1, l2)
+	// Empty strings always compare full equal
+	if maxLen == 0 {
+		return 1.0
+	}
+	dist := levenshtein.ComputeDistance(s1, s2)
+	// fmt.Printf("%v (%v) ~ %v (%v) = %v\n", s1, l1, s2, l2, dist)
+	return 1.0 - (float64(dist) / float64(maxLen))
+}
+
+var reExtraTitleInfo = regexp.MustCompile(`\([^)]+\)$`)
+var reMultiSpace = regexp.MustCompile(`\s+`)
+
+// Normalizes a track or release title.
+func NormalizeTitle(s string) string {
+	s = strings.TrimSpace(s)
+	s = strings.ToLower(s)
+	s = reExtraTitleInfo.ReplaceAllString(s, "")
+	s = reMultiSpace.ReplaceAllString(s, " ")
+	return s
+}

internal/similarity/similarity_test.go

@@ -0,0 +1,51 @@
+/*
+Copyright © 2024 Philipp Wolfer <phw@uploadedlobster.com>
+
+Scotty is free software: you can redistribute it and/or modify it under the
+terms of the GNU General Public License as published by the Free Software
+Foundation, either version 3 of the License, or (at your option) any later version.
+
+Scotty is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
+A PARTICULAR PURPOSE. See the GNU General Public License for more details.
+
+You should have received a copy of the GNU General Public License along with
+Scotty. If not, see <https://www.gnu.org/licenses/>.
+*/
+
+package similarity_test
+
+import (
+	"fmt"
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+	"go.uploadedlobster.com/scotty/internal/similarity"
+)
+
+func ExampleSimilarity() {
+	s := similarity.Similarity("bar1", "bär1")
+	fmt.Println(s)
+	// Output: 0.75
+}
+
+func TestSimilarity(t *testing.T) {
+	assert := assert.New(t)
+	assert.Equal(1.0, similarity.Similarity("", ""))
+	assert.Equal(0.0, similarity.Similarity("foo", ""))
+	assert.Equal(0.0, similarity.Similarity("foo", "bar"))
+	assert.Equal(0.5, similarity.Similarity("foobar", "bar"))
+	assert.Equal(1.0, similarity.Similarity("foo", "foo"))
+}
+
+func ExampleNormalizeTitle() {
+	s := similarity.NormalizeTitle(" Forever \tFailure (video edit) ")
+	fmt.Println(s)
+	// Output: forever failure
+}
+
+func TestNormalizeTitle(t *testing.T) {
+	assert := assert.New(t)
+	assert.Equal("forever failure", similarity.NormalizeTitle("Forever Failure"))
+	assert.Equal("foo", similarity.NormalizeTitle(" \tfoo\t \t"))
+}