Last active
November 27, 2016 04:31
-
-
Save tevjef/cf451c3947191299372879fd08c1e644 to your computer and use it in GitHub Desktop.
Golang regexp vs strings.Map
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
bash-3.2$ go test -bench=. | |
PASS | |
BenchmarkRegexReplace-4 500000 2470 ns/op 19.83 MB/s 113 B/op 1 allocs/op | |
BenchmarkMapReplace-4 2000000 813 ns/op 60.22 MB/s 0 B/op 0 allocs/op |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
package main | |
import ( | |
"regexp" | |
"strings" | |
"unicode" | |
) | |
var topicRegex, _ = regexp.Compile("[^A-Za-z0-9-_.~% ]+") | |
func regexReplace(str string) string { | |
return topicRegex.ReplaceAllLiteralString(str, "") | |
} | |
func mapReplace(str string) string { | |
return strings.Map(func(r rune) rune { | |
if isValidTopicRune(r) { | |
return r | |
} | |
return -1 | |
}, str) | |
} | |
func isValidTopicRune(r rune) bool { | |
isDash := r == rune('-') | |
isUnderscore := r == rune('_') | |
isDot := r == rune('.') | |
isTilde := r == rune('~') | |
isPercent := r == rune('%') | |
return unicode.IsLetter(r) || unicode.IsNumber(r) || unicode.IsSpace(r) || isDash || isUnderscore || isDot || isTilde || isPercent | |
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
package main | |
import ( | |
"math/rand" | |
"testing" | |
"time" | |
) | |
var testString = "Res Proposal In A....H.!@#$%[]^&*()_?><.02.87A\u0000SDA" | |
var expected = "Res Proposal In A....H.%_.02.87ASDA" | |
func Test_regexReplace(t *testing.T) { | |
type args struct { | |
str string | |
} | |
tests := []struct { | |
name string | |
args args | |
want string | |
}{ | |
{args: args{testString}, want: expected}, | |
} | |
for _, tt := range tests { | |
if got := regexReplace(tt.args.str); got != tt.want { | |
t.Errorf("%q. regexReplace() = %v, want %v", tt.name, got, tt.want) | |
} | |
} | |
} | |
func Test_mapReplace(t *testing.T) { | |
type args struct { | |
str string | |
} | |
tests := []struct { | |
name string | |
args args | |
want string | |
}{ | |
{args: args{testString}, want: expected}, | |
} | |
for _, tt := range tests { | |
if got := mapReplace(tt.args.str); got != tt.want { | |
t.Errorf("%q. mapReplace() = %v, want %v", tt.name, got, tt.want) | |
} | |
} | |
} | |
func BenchmarkRegexReplace(b *testing.B) { | |
seed := time.Now().UnixNano() | |
popr := rand.New(rand.NewSource(seed)) | |
total := 0 | |
pops := make([]string, 10000) | |
for i := 0; i < 10000; i++ { | |
pops[i] = randStringModel(popr) | |
} | |
b.ReportAllocs() | |
b.ResetTimer() | |
for i := 0; i < b.N; i++ { | |
total += len(regexReplace(pops[i%10000])) | |
} | |
b.SetBytes(int64(total / b.N)) | |
} | |
func BenchmarkMapReplace(b *testing.B) { | |
popr := rand.New(rand.NewSource(0)) | |
total := 0 | |
pops := make([]string, 10000) | |
for i := 0; i < 10000; i++ { | |
pops[i] = randStringModel(popr) | |
} | |
b.ReportAllocs() | |
b.ResetTimer() | |
for i := 0; i < b.N; i++ { | |
total += len(mapReplace(pops[i%10000])) | |
} | |
b.SetBytes(int64(total / b.N)) | |
} | |
type randyModel interface { | |
Float32() float32 | |
Float64() float64 | |
Int63() int64 | |
Int31() int32 | |
Uint32() uint32 | |
Intn(n int) int | |
} | |
func randStringModel(r randyModel) string { | |
v32 := r.Intn(100) | |
tmps := make([]rune, v32) | |
for i := 0; i < v32; i++ { | |
tmps[i] = randUTF8RuneModel(r) | |
} | |
return string(tmps) | |
} | |
func randUTF8RuneModel(r randyModel) rune { | |
ru := r.Intn(62) | |
if ru < 10 { | |
return rune(ru + 48) | |
} else if ru < 36 { | |
return rune(ru + 55) | |
} | |
return rune(ru + 61) | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment