1、使用 rune

func BenchmarkSubstr1(b *testing.B) {
	s := "bench基准测试,1212"
	for i := 0; i < b.N; i++ {
		s1 := []rune(s)
		s2 := s1[1:6]
		_ = string(s2)
	}
}

2、使用 utf 包

utf8.DecodeRuneInString()
func DecodeRuneInString(s string) (r rune, size int)

返回一个 rune 类型,以及此字符占用几个字节。

func BenchmarkSubstr2(b *testing.B) {
	s := "bench基准测试,1212"
	st := 1
	en := 6
	for i := 0; i < b.N; i++ {
		var start, end int
		var c int
		for i := 0; i <= en && c < len(s); i++ {
			_, size := utf8.DecodeRuneInString(s[c:])
			if i == st {
				start = c
			}
			if i == en {
				end = c
			}
			c += size
		}
		_ = s[start:end]
	}
}

3、使用 range 遍历

range字符ASCII中文字符
func TestSubstr(t *testing.T) {
	s := "bench基准测试,1212"
	for k, v := range s {
		fmt.Println(k, string(v))
	}
}

打印输出

0 b
1 e
2 n
3 c
4 h
5 基
8 准
11 测
14 试
17 ,
20 1
21 2
22 1
23 2

可见 range 默认就支持 unicode,因此变量 k 不是连续的,找到 k 即可。

func TestSubstr(t *testing.T) {
	s := "bench基准测试,1212"
	var start, end int
	var c int
	for k := range s {
		if c == 1 {
			start = k
		}
		if c == 6 {
			end = k
		}
		c++
	}
	fmt.Println(s[start:end]) // ench基
}
func BenchmarkSubstr3(b *testing.B) {
	s := "bench基准测试,1212"
	for i := 0; i < b.N; i++ {
		var start, end int
		var c int
		for k := range s {
			if c == 1 {
				start = k
			}
			if c == 6 {
				end = k
			}
			c++
		}
		_ = s[start:end]
	}
}
> go test -bench=BenchmarkSubstr -benchtime=100000x -benchmem
goos: windows
goarch: amd64
pkg: test
cpu: Intel(R) Core(TM) i7-4790 CPU @ 3.60GHz
BenchmarkSubstr1-8        100000               107.5 ns/op             0 B/op          0 allocs/op
BenchmarkSubstr2-8        100000                37.74 ns/op            0 B/op          0 allocs/op
BenchmarkSubstr3-8        100000                37.72 ns/op            0 B/op          0 allocs/op