新增FindBestMatch接口

Author: guonaihong

README.md

@@ -18,6 +18,7 @@ strsim是golang实现的字符串相识度库，后端集成多种算法，主
 ## 内容
 - [比较两个字符串相识度](#比较两个字符串相识度)
 - [从字符串数组里面找到相似度最高的字符串](#从数组里找到相似度最高的字符串)
+- [从字符串数组里面找到相似度最高的字符串-带下标](#从数组里找到相似度最高的字符串-带下标)
 - [选择不同算法](##选择不同算法)
     - [莱文斯坦-编辑距离(Levenshtein)](#莱文斯坦-编辑距离(Levenshtein))
     - [选择Dice's coefficient](#选择Dice's-coefficient)
@@ -33,6 +34,10 @@ strsim.Compare("中国人", "中")
 ```go
 strsim.FindBestMatchOne("海刘", []string{"白日依山尽", "黄河入海流", "欲穷千里目", "更上一层楼"})
 ```
+## 从数组里找到相似度最高的字符串-带下标
+```go
+strsim.FindBestMatch("海刘", []string{"白日依山尽", "黄河入海流", "欲穷千里目", "更上一层楼"})
+```
 
 ## 选择不同算法
 ### 莱文斯坦-编辑距离(Levenshtein)

strsim.go

@@ -19,46 +19,7 @@ func FindBestMatchOne(s string, targets []string, opts ...Option) *similarity.Ma
 	return r.Match
 }
 
-// 比较两个字符串内部函数
-func compare(s1, s2 string, o *option) float64 {
-	if s, e := modifyStrAndCheck(o, &s1, &s2); e {
-		return s
-	}
-
-	return o.cmp(s1, s2)
-}
-
-// 前处理主要涉及，修改字符串，和边界判断
-func modifyStrAndCheck(o *option, s1, s2 *string) (score float64, exit bool) {
-	modifyString(o, s1)
-	modifyString(o, s2)
-
-	return check(*s1, *s2)
-}
-
-// 记录每个targets子串的相似度打分，并且返回相似度最高的那个字符串, 内部函数
-func findBestMatch(s string, targets []string, opts ...Option) *similarity.MatchResult {
-
-	var opt option
-	opt.fillOption(opts...)
-
-	match := make([]*similarity.Match, 0, len(targets))
-	bestIndex := 0
-	for k, s2 := range targets {
-
-		score := compare(s, s2, &opt)
-
-		//fmt.Printf("score:%f(%s)(%s)\n", score, s, s2)
-		match = append(match, &similarity.Match{S: s2, Score: score})
-
-		if k == 0 {
-			continue
-		}
-
-		if score > match[bestIndex].Score {
-			bestIndex = k
-		}
-	}
-
-	return &similarity.MatchResult{AllResult: match, Match: match[bestIndex], BestIndex: bestIndex}
+// 返回相似度最高的那个字符串, 以及索引位置
+func FindBestMatch(s string, targets []string, opts ...Option) *similarity.MatchResult {
+	return findBestMatch(s, targets, opts...)
 }

strsim_priv.go

@@ -0,0 +1,47 @@
+package strsim
+
+import "github.com/antlabs/strsim/similarity"
+
+// 比较两个字符串内部函数
+func compare(s1, s2 string, o *option) float64 {
+	if s, e := modifyStrAndCheck(o, &s1, &s2); e {
+		return s
+	}
+
+	return o.cmp(s1, s2)
+}
+
+// 前处理主要涉及，修改字符串，和边界判断
+func modifyStrAndCheck(o *option, s1, s2 *string) (score float64, exit bool) {
+	modifyString(o, s1)
+	modifyString(o, s2)
+
+	return check(*s1, *s2)
+}
+
+// 记录每个targets子串的相似度打分，并且返回相似度最高的那个字符串, 内部函数
+func findBestMatch(s string, targets []string, opts ...Option) *similarity.MatchResult {
+
+	var opt option
+	opt.fillOption(opts...)
+
+	match := make([]*similarity.Match, 0, len(targets))
+	bestIndex := 0
+	for k, s2 := range targets {
+
+		score := compare(s, s2, &opt)
+
+		//fmt.Printf("score:%f(%s)(%s)\n", score, s, s2)
+		match = append(match, &similarity.Match{S: s2, Score: score})
+
+		if k == 0 {
+			continue
+		}
+
+		if score > match[bestIndex].Score {
+			bestIndex = k
+		}
+	}
+
+	return &similarity.MatchResult{AllResult: match, Match: match[bestIndex], BestIndex: bestIndex}
+}

strsim_test.go

@@ -45,3 +45,18 @@ func Test_FindBestMatchOne(t *testing.T) {
 		}
 	}
 }
+
+func Test_FindBestMatch(t *testing.T) {
+	for _, d := range []bestTest{
+		{best: []string{"朝辞白帝彩云间", "千里江陵一日还", "两岸猿声啼不住", "轻舟已过万重山"}, key: "千里还", need: "千里江陵一日还"},
+	} {
+		for _, o := range []Option{
+			DiceCoefficient(1),
+			Jaro(),
+			Default(),
+		} {
+			m := FindBestMatch(d.key, d.best, o)
+			assert.Equal(t, m.Match.S, d.need)
+		}
+	}
+}