Fuzzy search retruning set of matching set (optinnal threshold) + tests

hbollon · Sep 15, 2020 · 1acfbbe · 1acfbbe
1 parent b471431
commit 1acfbbe
Show file tree

Hide file tree

Showing 2 changed files with 155 additions and 1 deletion.
diff --git a/string-analysis.go b/string-analysis.go
@@ -2,7 +2,10 @@ package edlib
 
 import (
 	"errors"
+	"fmt"
 	"log"
+
+	"github.com/hbollon/go-edlib/internal/orderedmap"
 )
 
 // AlgorithMethod is an Integer type used to identify edit distance algorithms
@@ -77,7 +80,7 @@ func FuzzySearch(str string, strList []string, algo AlgorithMethod) string {
 }
 
 // FuzzySearchThreshold realize an approximate search on a string list and return the closest one compared
-// to the string input. Take an similarity threshold in parameter.
+// to the string input. Takes a similarity threshold in parameter.
 func FuzzySearchThreshold(str string, strList []string, minSim float32, algo AlgorithMethod) string {
 	var higherMatchPercent float32
 	var tmpStr string
@@ -96,3 +99,46 @@ func FuzzySearchThreshold(str string, strList []string, minSim float32, algo Alg
 	}
 	return tmpStr
 }
+
+// FuzzySearchSet realize an approximate search on a string list and return a set composed with x strings compared
+// to the string input sorted by similarity with the base string. Takes the a quantity parameter to define the number of output strings desired (For exemple 3 in the case of the Google Keyborad word suggestion).
+func FuzzySearchSet(str string, strList []string, quantity int, algo AlgorithMethod) []string {
+	sortedMap := make(orderedmap.OrderedMap, quantity)
+	for _, strToCmp := range strList {
+		sim, err := StringsSimilarity(str, strToCmp, algo)
+		if err != nil {
+			log.Fatal(err)
+		} else {
+			fmt.Printf("Sim %s/%s : %f\n", str, strToCmp, sim)
+		}
+
+		if sim > sortedMap[sortedMap.Len()-1].Value {
+			sortedMap[sortedMap.Len()-1].Key = strToCmp
+			sortedMap[sortedMap.Len()-1].Value = sim
+			sortedMap.SortByValues()
+		}
+	}
+
+	return sortedMap.ToArray()
+}
+
+// FuzzySearchSetThreshold realize an approximate search on a string list and return a set composed with x strings compared
+// to the string input sorted by similarity with the base string. Take a similarity threshold in parameter. Takes the a quantity parameter to define the number of output strings desired (For exemple 3 in the case of the Google Keyborad word suggestion).
+// Takes also a threshold parameter for similarity with base string.
+func FuzzySearchSetThreshold(str string, strList []string, quantity int, minSim float32, algo AlgorithMethod) []string {
+	sortedMap := make(orderedmap.OrderedMap, quantity)
+	for _, strToCmp := range strList {
+		sim, err := StringsSimilarity(str, strToCmp, algo)
+		if err != nil {
+			log.Fatal(err)
+		}
+
+		if sim >= minSim && sim > sortedMap[sortedMap.Len()-1].Value {
+			sortedMap[sortedMap.Len()-1].Key = strToCmp
+			sortedMap[sortedMap.Len()-1].Value = sim
+			sortedMap.SortByValues()
+		}
+	}
+
+	return sortedMap.ToArray()
+}
diff --git a/tests/string-analysis_test.go b/tests/string-analysis_test.go
@@ -1,11 +1,26 @@
 package edlib
 
 import (
+	"reflect"
 	"testing"
 
 	"github.com/hbollon/go-edlib"
 )
 
+var strList []string
+
+func init() {
+	strList = []string{
+		"test",
+		"tester",
+		"tests",
+		"testers",
+		"testing",
+		"tsting",
+		"sting",
+	}
+}
+
 func TestStringsSimilarity(t *testing.T) {
 	type args struct {
 		str1 string
@@ -118,3 +133,96 @@ func TestStringsSimilarity(t *testing.T) {
 		})
 	}
 }
+
+func TestFuzzySearch(t *testing.T) {
+	type args struct {
+		str     string
+		strList []string
+		algo    edlib.AlgorithMethod
+	}
+	tests := []struct {
+		name string
+		args args
+		want string
+	}{
+		{"FuzzySearch 'testing'", args{"testnig", strList, edlib.Levenshtein}, "testing"},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := edlib.FuzzySearch(tt.args.str, tt.args.strList, tt.args.algo); got != tt.want {
+				t.Errorf("FuzzySearch() = %v, want %v", got, tt.want)
+			}
+		})
+	}
+}
+
+func TestFuzzySearchThreshold(t *testing.T) {
+	type args struct {
+		str     string
+		strList []string
+		minSim  float32
+		algo    edlib.AlgorithMethod
+	}
+	tests := []struct {
+		name string
+		args args
+		want string
+	}{
+		{"FuzzySearch 'testing'", args{"testnig", strList, 0.7, edlib.Levenshtein}, "testing"},
+		{"FuzzySearch 'testing'", args{"hello", strList, 0.7, edlib.Levenshtein}, ""},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := edlib.FuzzySearchThreshold(tt.args.str, tt.args.strList, tt.args.minSim, tt.args.algo); got != tt.want {
+				t.Errorf("FuzzySearchThreshold() = %v, want %v", got, tt.want)
+			}
+		})
+	}
+}
+
+func TestFuzzySearchSet(t *testing.T) {
+	type args struct {
+		str      string
+		strList  []string
+		quantity int
+		algo     edlib.AlgorithMethod
+	}
+	tests := []struct {
+		name string
+		args args
+		want []string
+	}{
+		{"FuzzySearch 'testing'", args{"testnig", strList, 3, edlib.Levenshtein}, []string{"testing", "test", "tester"}},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := edlib.FuzzySearchSet(tt.args.str, tt.args.strList, tt.args.quantity, tt.args.algo); !reflect.DeepEqual(got, tt.want) {
+				t.Errorf("FuzzySearchSet() = %v, want %v", got, tt.want)
+			}
+		})
+	}
+}
+
+func TestFuzzySearchSetThreshold(t *testing.T) {
+	type args struct {
+		str      string
+		strList  []string
+		quantity int
+		minSim   float32
+		algo     edlib.AlgorithMethod
+	}
+	tests := []struct {
+		name string
+		args args
+		want []string
+	}{
+		{"FuzzySearch 'testing'", args{"testnig", strList, 3, 0.7, edlib.Levenshtein}, []string{"testing", "", ""}},
+	}
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			if got := edlib.FuzzySearchSetThreshold(tt.args.str, tt.args.strList, tt.args.quantity, tt.args.minSim, tt.args.algo); !reflect.DeepEqual(got, tt.want) {
+				t.Errorf("FuzzySearchSetThreshold() = %v, want %v", got, tt.want)
+			}
+		})
+	}
+}