golang版本快速排序实现

quicksort简介

使用快速排序算法，对slice进行排序。目前支持int，int8，int16，int32，int64，float32，float64，string，uint，uint8，uint16，uint32，uint64类型的slice

连接地址：github上quicksort源码地址

go get github.com/hzwy23/quicksort

使用例子

package mainimport (    "fmt"    "time"    "github.com/hzwy23/quicksort")func main() {    var t2 []int = []int{11, 31, 43, 3, 48, 23}    quicksort.QuickSort(t2)    fmt.Println("int", t2)    var t28 []int8 = []int8{11, 31, 43, 3, 48, 23}    quicksort.QuickSort(t28)    fmt.Println("int8", t28)    var t216 []int16 = []int16{11, 31, 43, 3, 48, 23}    quicksort.QuickSort(t216)    fmt.Println("int16", t216)    var t232 []int32 = []int32{11, 31, 43, 3, 48, 23}    quicksort.QuickSort(t232)    fmt.Println("int32", t232)    var t264 []int64 = []int64{11, 31, 43, 3, 48, 23}    quicksort.QuickSort(t264)    fmt.Println("int64", t264)    var t3 []string = []string{"123", "211", "abc", "def", "1sd", "6dw", "235"}    quicksort.QuickSort(t3)    fmt.Println("string", t3)    var tf32 []float32 = []float32{3.2345, 1.242345, 3.11234, 6.41234, 4.11235}    quicksort.QuickSort(tf32)    fmt.Println("float32", tf32)    var t4 []float64 = []float64{3.2345, 1.242345, 3.11234, 6.41234, 4.11235}    quicksort.QuickSort(t4)    fmt.Println("float64", t4)    var t8 []int8 = []int8{19, 43, 63, 27, 41, 24, 64, 24, 64, 34, 65, 24, 6, 1, 23, 5, 43, 6, -1, -5, -2, 23, 4}    quicksort.QuickSort(t8)    fmt.Println("t8", t8)    var t16 []int16 = []int16{19, 43, 63, 27, 41, 24, 64, 24, 64, 234, 645, 234, 6, 1, 23, 5, 43, 6, -1, -5, -2, 4123, 4234}    quicksort.QuickSort(t16)    fmt.Println("t16", t16)    var t32 []int32 = []int32{19, 43, 63, 27, 41, 24, 64, 24, 64, 234, 64345, 234, 6, 1, 23, 5, 43, 6, -1, -5, -2, 4123, 4234}    quicksort.QuickSort(t32)    fmt.Println("t32", t32)    var t64 []int64 = []int64{19, 43, 63, 27, 41, 24, 64, 24, 64, 234, 64345, 234, 6, 1, 23, 5, 43, 6, -1, -5, -2, 4123, 4234}    quicksort.QuickSort(t64)    fmt.Println("t64", t64)    var utint []uint = []uint{12, 3, 4, 5}    quicksort.QuickSort(utint)    fmt.Println("utint", utint)    var utint8 []uint8 = []uint8{12, 3, 4, 5}    quicksort.QuickSort(utint8)    fmt.Println("utint", utint8)    var utint16 []uint16 = []uint16{12, 3, 4, 5}    quicksort.QuickSort(utint16)    fmt.Println("utint", utint16)    var utint32 []uint32 = []uint32{12, 3, 4, 5}    quicksort.QuickSort(utint32)    fmt.Println("utint", utint32)    var utint64 []uint64 = []uint64{12, 3, 4, 5}    quicksort.QuickSort(utint64)    fmt.Println("utint", utint64)}

通过反射方式，自动匹配slice类型

性能介绍

golang sdk包中已经实现了quicksort，但是在使用时，需要先转换一次，才能使用sdk中提供的quicksort方法。在处理不重复的slice时，在指定随机的slice数据测试中，sdk的quicksort性能比我这边提供的包性能稍逊一些，所以，在此贴出我的quicksort实现方式。不足之处，请各位指正与谅解。下边附上测试代码：

func TestSysSort(t *testing.T) {    fmt.Println("start system sort")    curTime := time.Now().Unix()    fmt.Println(time.Now().Format("2006-01-02 15:04:05"))    for i := 0; i < 10000000; i++ {        var t []int = []int{11, -1, 3, 4, 5, 12, 545, 67456, 34, 6, 3, 45, 7, 3, 5, 5, 6, 3, 3, 4, 7, 764, 465, 5423}        a := sort.IntSlice(t[0:])        sort.Sort(a)    }    fmt.Println(time.Now().Format("2006-01-02 15:04:05"))    endTime := time.Now().Unix()    fmt.Println("end system sort,costs time:", (endTime - curTime))}func TestMySort(t *testing.T) {    fmt.Println("start my sort")    curTime := time.Now().Unix()    fmt.Println(time.Now().Format("2006-01-02 15:04:05"))    for i := 0; i < 10000000; i++ {        var t []int = []int{11, -1, 3, 4, 5, 12, 545, 67456, 34, 6, 3, 45, 7, 3, 5, 5, 6, 3, 3, 4, 7, 764, 465, 5423}        quicksort.QuickSort(t)    }    fmt.Println(time.Now().Format("2006-01-02 15:04:05"))    endTime := time.Now().Unix()    fmt.Println("end my sort,costs time:", (endTime - curTime))}

上述是测试代码，请自行测试，对比性能。