strings: better mean complexity for Count and Index.

The O(n+m) complexity is obtained probabilistically
by using Rabin-Karp algorithm, which provides the needed complexity
unless exceptional collisions occur, without memory allocation.

benchmark                 old ns/op    new ns/op    delta
BenchmarkIndexHard1         6532331      4045886  -38.06%
BenchmarkIndexHard2         8178173      4038975  -50.61%
BenchmarkIndexHard3         6973687      4042591  -42.03%
BenchmarkCountHard1         6270864      4071090  -35.08%
BenchmarkCountHard2         7838039      4072853  -48.04%
BenchmarkCountHard3         6697828      4071964  -39.20%
BenchmarkIndexTorture       2730546        28934  -98.94%
BenchmarkCountTorture       2729622        29064  -98.94%

Fixes #4600.

R=rsc, donovanhide, remyoudompheng
CC=golang-dev
https://golang.org/cl/7314095
This commit is contained in:
Rémy Oudompheng 2013-02-17 13:07:17 +01:00
parent f1037f0e86
commit 23093f86ee
2 changed files with 109 additions and 8 deletions

View file

@ -1001,6 +1001,57 @@ func TestEqualFold(t *testing.T) {
}
}
func makeBenchInputHard() string {
tokens := [...]string{
"<a>", "<p>", "<b>", "<strong>",
"</a>", "</p>", "</b>", "</strong>",
"hello", "world",
}
x := make([]byte, 0, 1<<20)
for len(x) < 1<<20 {
i := rand.Intn(len(tokens))
x = append(x, tokens[i]...)
}
return string(x)
}
var benchInputHard = makeBenchInputHard()
func benchmarkIndexHard(b *testing.B, sep string) {
for i := 0; i < b.N; i++ {
Index(benchInputHard, sep)
}
}
func benchmarkCountHard(b *testing.B, sep string) {
for i := 0; i < b.N; i++ {
Count(benchInputHard, sep)
}
}
func BenchmarkIndexHard1(b *testing.B) { benchmarkIndexHard(b, "<>") }
func BenchmarkIndexHard2(b *testing.B) { benchmarkIndexHard(b, "</pre>") }
func BenchmarkIndexHard3(b *testing.B) { benchmarkIndexHard(b, "<b>hello world</b>") }
func BenchmarkCountHard1(b *testing.B) { benchmarkCountHard(b, "<>") }
func BenchmarkCountHard2(b *testing.B) { benchmarkCountHard(b, "</pre>") }
func BenchmarkCountHard3(b *testing.B) { benchmarkCountHard(b, "<b>hello world</b>") }
var benchInputTorture = Repeat("ABC", 1<<10) + "123" + Repeat("ABC", 1<<10)
var benchNeedleTorture = Repeat("ABC", 1<<10+1)
func BenchmarkIndexTorture(b *testing.B) {
for i := 0; i < b.N; i++ {
Index(benchInputTorture, benchNeedleTorture)
}
}
func BenchmarkCountTorture(b *testing.B) {
for i := 0; i < b.N; i++ {
Count(benchInputTorture, benchNeedleTorture)
}
}
var makeFieldsInput = func() string {
x := make([]byte, 1<<20)
// Input is ~10% space, ~10% 2-byte UTF-8, rest ASCII non-space.