golang 正则匹配regexp接口实战学习

总览

如果不熟悉正则表达式的语法的话，可以执行下面命令：

go doc regexp/syntax

联系代码

package mainimport (    "fmt"    "regexp")func expandTest() {    pat := `(((abc.)def.)ghi)`    reg := regexp.MustCompile(pat)    fmt.Println(reg.NumSubexp())    src := []byte(`abc-def-ghi abc+def+ghi`)    template := []byte(`$0   $1   $2   $3`)    // 替换第一次匹配结果    match := reg.FindSubmatchIndex(src)    fmt.Printf("%v\n", match) // [0 11 0 11 0 8 0 4]    dst := reg.Expand(nil, template, src, match)    fmt.Printf("%s\n\n", dst)    // abc-def-ghi   abc-def-ghi   abc-def-   abc-    // 替换所有匹配结果    for _, match := range reg.FindAllSubmatchIndex(src, -1) {        fmt.Printf("%v\n", match)        dst := reg.Expand(nil, template, src, match)        fmt.Printf("%s\n", dst)    }    // [0 11 0 11 0 8 0 4]    // abc-def-ghi   abc-def-ghi   abc-def-   abc-    // [12 23 12 23 12 20 12 16]    // abc+def+ghi   abc+def+ghi   abc+def+   abc+}func testFind() {    re := regexp.MustCompile("a*r")    fmt.Println(string(re.Find([]byte("paranoabrmal"))))    fmt.Println(re.NumSubexp())    rep := regexp.MustCompilePOSIX("a*r|ara")    fmt.Println(string(rep.Find([]byte("paranoabrmal"))))    fmt.Println(rep.NumSubexp())    b := []byte("abc1def1")    pat := `abc1|abc1def1`    reg1 := regexp.MustCompile(pat)      // 第一匹配    reg2 := regexp.MustCompilePOSIX(pat) // 最长匹配    fmt.Printf("%s\n", reg1.Find(b))     // abc1    fmt.Printf("%s\n", reg2.Find(b))     // abc1def1    fmt.Println(reg1.NumSubexp())    b = []byte("abc1def1")    pat = `(abc|abc1def)*1`    reg1 = regexp.MustCompile(pat)      // 第一匹配    reg2 = regexp.MustCompilePOSIX(pat) // 最长匹配    fmt.Printf("%s\n", reg1.Find(b))    // abc1    fmt.Printf("%s\n", reg2.Find(b))    // abc1def1    fmt.Println(reg1.NumSubexp())}func testFindAll() {    re := regexp.MustCompile("ar")    fmt.Printf("%q\n", (re.FindAll([]byte("paranoarmal"), -1)))    rep := regexp.MustCompilePOSIX("ar")    fmt.Printf("%q\n", (rep.FindAll([]byte("paranoarmal"), -1)))    pat := `(((abc.)def.)ghi)`    src := []byte(`abc-def-ghi abc+def+ghi`)    reg := regexp.MustCompile(pat)    fmt.Printf("%q\n", (reg.Find(src)))    fmt.Printf("%q\n", (reg.FindAll(src, -1)))    regp := regexp.MustCompilePOSIX(pat)    fmt.Printf("%q\n", (regp.Find(src)))    fmt.Printf("%q\n", (regp.FindAll(src, -1)))}func testReg(pat, srcStr string) {    fmt.Println("xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx")    src := []byte(srcStr)    reg := regexp.MustCompile(pat)    fmt.Printf("%q\n", (reg.Find(src)))    fmt.Printf("%q\n", (reg.FindString(srcStr)))    fmt.Printf("%q\n", (reg.FindAll(src, -1)))    fmt.Printf("%q\n", (reg.FindAllString(srcStr, -1)))    fmt.Printf("%d\n", (reg.FindIndex(src)))    fmt.Printf("%d\n", (reg.FindStringIndex(srcStr)))    fmt.Printf("%d\n", (reg.FindAllIndex(src, -1)))    fmt.Printf("%d\n", (reg.FindAllStringIndex(srcStr, -1)))    fmt.Println("begin submatch")    fmt.Printf("%q\n", (reg.FindSubmatch(src)))    fmt.Printf("%q\n", (reg.FindStringSubmatch(srcStr)))    fmt.Printf("%d\n", (reg.FindSubmatchIndex(src)))    fmt.Printf("%d\n", (reg.FindStringSubmatchIndex(srcStr)))    fmt.Printf("%q\n", (reg.FindAllSubmatch(src, -1)))    fmt.Printf("%q\n", (reg.FindAllStringSubmatch(srcStr, -1)))    fmt.Printf("%d\n", (reg.FindAllSubmatchIndex(src, -1)))    fmt.Printf("%d\n", (reg.FindAllStringSubmatchIndex(srcStr, -1)))    regp := regexp.MustCompilePOSIX(pat)    fmt.Printf("%q\n", (regp.Find(src)))    fmt.Printf("%q\n", (regp.FindString(srcStr)))    fmt.Printf("%q\n", (regp.FindAll(src, -1)))    fmt.Printf("%q\n", (regp.FindAllString(srcStr, -1)))    fmt.Printf("%d\n", (regp.FindIndex(src)))    fmt.Printf("%d\n", (regp.FindStringIndex(srcStr)))    fmt.Printf("%d\n", (regp.FindAllIndex(src, -1)))    fmt.Printf("%d\n", (regp.FindAllStringIndex(srcStr, -1)))    fmt.Println("begin submatch")    fmt.Printf("%q\n", (regp.FindSubmatch(src)))    fmt.Printf("%q\n", (regp.FindStringSubmatch(srcStr)))    fmt.Printf("%q\n", (regp.FindAllSubmatch(src, -1)))    fmt.Printf("%q\n", (regp.FindAllStringSubmatch(srcStr, -1)))    fmt.Printf("%d\n", (regp.FindSubmatchIndex(src)))    fmt.Printf("%d\n", (regp.FindStringSubmatchIndex(srcStr)))    fmt.Printf("%d\n", (regp.FindAllSubmatchIndex(src, -1)))    fmt.Printf("%d\n", (regp.FindAllStringSubmatchIndex(srcStr, -1)))}func main() {    testReg(`(((abc.)def.)ghi)x*`, `abc-def-ghixxa abc+def+ghixx`)    testReg(`(((abc.)def.)ghi)`, `abc-def-ghixxa abc+def+ghixx`)    testReg(`a(x*)b(y|z)c`, `-axxxbyc- -abzc-`)    testReg(`a*r`, `paranoabrmal`)}

代码详解

Find FindString

fmt.Printf("%q\n", (reg.Find(src)))fmt.Printf("%q\n", (reg.FindString(srcStr)))

Find返回leftmost的正则匹配结果，也就是满足匹配的最左边的最短匹配字符串

testReg(`(((abc.)def.)ghi)x*`, `abc-def-ghixxa abc+def+ghixx`)结果："abc-def-ghixx"分析：*表示0或多个，优先选择多个，所以优先匹配为"abc-def-ghixx"testReg(`(((abc.)def.)ghi)`, `abc-def-ghixxa abc+def+ghixx`)结果："abc-def-ghi"分析：满足匹配项的的最左的最短匹配就是"abc-def-ghi"testReg(`a(x*)b(y|z)c`, `-axxxbyc- -abzc-`)结果："axxxbyc"分析：满足匹配项的的最左的最短匹配就是"abc-def-ghi"testReg(`a*r`, `paranoabrmal`)结果："ar"分析：满足匹配项的的最左的最短匹配就是"ar"

FindAll FIndAllString

fmt.Printf("%q\n", (reg.FindAll(src, -1)))fmt.Printf("%q\n", (reg.FindAllString(srcStr, -1)))

FindAll是Find的“ALL”版本，返回相邻的非互相覆盖的最短匹配项

testReg(`(((abc.)def.)ghi)x*`, `abc-def-ghixxa abc+def+ghixx`)结果：["abc-def-ghixx" "abc+def+ghixx"]分析：返回所有匹配项testReg(`(((abc.)def.)ghi)`, `abc-def-ghixxa abc+def+ghixx`)结果：["abc-def-ghi" "abc+def+ghi"]分析：返回所有匹配项testReg(`a(x*)b(y|z)c`, `-axxxbyc- -abzc-`)结果：["axxxbyc" "abzc"]分析：返回所有匹配项testReg(`a*r`, `paranoabrmal`)结果：["ar" "r"]分析：*表示0或多个，所以第二个就仅仅匹配了r

FindIndex FindStringIndex

fmt.Printf("%d\n", (reg.FindIndex(src)))fmt.Printf("%d\n", (reg.FindStringIndex(srcStr)))

FindIndex是Find的“Index”版本，匹配项在原字符串中的位置，返回nil标示没有找到匹配项。

testReg(`(((abc.)def.)ghi)x*`, `abc-def-ghixxa abc+def+ghixx`)结果：[0 13]分析：`abc-def-ghixxa abc+def+ghixx`[0:13]就是"abc-def-ghixx"testReg(`(((abc.)def.)ghi)`, `abc-def-ghixxa abc+def+ghixx`)结果：[0 11]分析：同上testReg(`a(x*)b(y|z)c`, `-axxxbyc- -abzc-`)结果：[1 8]分析：同上testReg(`a*r`, `paranoabrmal`)结果：[1 3]分析：同上

FindAllIndex FindAllStringIndex

fmt.Printf("%d\n", (reg.FindAllIndex(src, -1)))fmt.Printf("%d\n", (reg.FindAllStringIndex(srcStr, -1)))

FindAllIndex是FindAll的“Index”版本，标示匹配项在原字符串中的位置，返回nil标示没有找到匹配项。

testReg(`(((abc.)def.)ghi)x*`, `abc-def-ghixxa abc+def+ghixx`)结果：[[0 13] [15 28]]分析：`abc-def-ghixxa abc+def+ghixx`[0:13]就是"abc-def-ghixx"        `abc-def-ghixxa abc+def+ghixx`[15:28]就是"abc+def+ghixx"testReg(`(((abc.)def.)ghi)`, `abc-def-ghixxa abc+def+ghixx`)结果：[[0 11] [15 26]]分析：同上testReg(`a(x*)b(y|z)c`, `-axxxbyc- -abzc-`)结果：[[1 8] [11 15]]分析：同上testReg(`a*r`, `paranoabrmal`)结果：[[1 3] [8 9]]分析：同上

FindSubmatch FindStringSubmatch

fmt.Printf("%q\n", (reg.FindSubmatch(src)))fmt.Printf("%q\n", (reg.FindStringSubmatch(srcStr)))

FindSubmatch是Find的“Submatch”版本，这个版本返回正则表达式的子匹配项。子匹配指的是在括号内的子正则表达式，从左到右进行标号。
子匹配0标示全部的表达式，子匹配1标示第1个括号内的子表达式，以此类推。

testReg(`(((abc.)def.)ghi)x*`, `abc-def-ghixxa abc+def+ghixx`)结果：["abc-def-ghixx" "abc-def-ghi" "abc-def-" "abc-"]分析：“.”匹配任意字符。submatch0为 `(((abc.)def.)ghi)x*`,submatch1为`(((abc.)def.)ghi)`，submatch2为`((abc.)def.)`，submatch3为`(abc.)`，testReg(`(((abc.)def.)ghi)`, `abc-def-ghixxa abc+def+ghixx`)结果：["abc-def-ghi" "abc-def-ghi" "abc-def-" "abc-"]分析：“.”匹配任意字符。submatch0为 `(((abc.)def.)ghi)`,submatch1为`(((abc.)def.)ghi)`，submatch2为`((abc.)def.)`，submatch3为`(abc.)`，testReg(`a(x*)b(y|z)c`, `-axxxbyc- -abzc-`)结果：["axxxbyc" "xxx" "y"]分析：submatch0为`a(x*)b(y|z)c`，submatch1为`(x*)`，submatch2为`(y|z)`testReg(`a*r`, `paranoabrmal`)结果：["ar"]分析：submatch0为`a*r`

FindSubmatchIndex FindStringSubmatchIndex

fmt.Printf("%d\n", (reg.FindSubmatchIndex(src)))fmt.Printf("%d\n", (reg.FindStringSubmatchIndex(srcStr)))

FindSubmatchIndex是FindSubmatch的Index版本，返回匹配项的首尾下标

testReg(`(((abc.)def.)ghi)x*`, `abc-def-ghixxa abc+def+ghixx`)结果：[0 13 0 11 0 8 0 4]分析： `abc-def-ghixxa abc+def+ghixx`[0:13]="abc-def-ghixx"        `abc-def-ghixxa abc+def+ghixx`[0:11]="abc-def-ghi"        `abc-def-ghixxa abc+def+ghixx`[0:8]= "abc-def-"        `abc-def-ghixxa abc+def+ghixx`[0:4]="abc-"testReg(`(((abc.)def.)ghi)`, `abc-def-ghixxa abc+def+ghixx`)结果：[0 11 0 11 0 8 0 4]分析：同上testReg(`a(x*)b(y|z)c`, `-axxxbyc- -abzc-`)结果：[1 8 2 5 6 7]分析：同上testReg(`a*r`, `paranoabrmal`)结果：[1 3]分析：同上

FindAllSubmatch FindAllStringSubmatch

fmt.Printf("%q\n", (reg.FindAllSubmatch(src, -1)))fmt.Printf("%q\n", (reg.FindAllStringSubmatch(srcStr, -1)))

FindAllSubmatch是FindSubmatch的“All”版本，这个版本返回全部匹配项的子匹配项。
子匹配0标示全部的表达式，子匹配1标示第1个括号内的子表达式，以此类推。

testReg(`(((abc.)def.)ghi)x*`, `abc-def-ghixxa abc+def+ghixx`)结果：[["abc-def-ghixx" "abc-def-ghi" "abc-def-" "abc-"] ["abc+def+ghixx" "abc+def+ghi" "abc+def+" "abc+"]]分析：参考FindSubmatchtestReg(`(((abc.)def.)ghi)`, `abc-def-ghixxa abc+def+ghixx`)结果：[["abc-def-ghi" "abc-def-ghi" "abc-def-" "abc-"] ["abc+def+ghi" "abc+def+ghi" "abc+def+" "abc+"]]分析：参考FindSubmatchtestReg(`a(x*)b(y|z)c`, `-axxxbyc- -abzc-`)结果：[["axxxbyc" "xxx" "y"] ["abzc" "" "z"]]分析：参考FindSubmatchtestReg(`a*r`, `paranoabrmal`)结果：[["ar"] ["r"]]分析：参考FindSubmatch

FindAllSubmatchIndex FindAllStringSubmatchIndex

fmt.Printf("%d\n", (reg.FindAllSubmatchIndex(src, -1)))fmt.Printf("%d\n", (reg.FindAllStringSubmatchIndex(srcStr, -1)))

FindAllSubmatchIndex是FindAllSubmatch的Index版本，返回匹配项的首尾下标

testReg(`(((abc.)def.)ghi)x*`, `abc-def-ghixxa abc+def+ghixx`)结果：[[0 13 0 11 0 8 0 4] [15 28 15 26 15 23 15 19]]分析： 参考FindSubmatchIndextestReg(`(((abc.)def.)ghi)`, `abc-def-ghixxa abc+def+ghixx`)结果：[[0 11 0 11 0 8 0 4] [15 26 15 26 15 23 15 19]]分析： 参考FindSubmatchIndextestReg(`a(x*)b(y|z)c`, `-axxxbyc- -abzc-`)结果：[[1 8 2 5 6 7] [11 15 12 12 13 14]]分析： 参考FindSubmatchIndextestReg(`a*r`, `paranoabrmal`)结果：[[1 3] [8 9]]分析： 参考FindSubmatchIndex

测试代码参考了这篇文章