bytes, strings: optimize Split*

The relevant benchmark results on linux/amd64:

bytes:

SplitSingleByteSeparator-4   25.7ms ± 5%   9.1ms ± 4%  -64.40%  (p=0.000 n=10+10)
SplitMultiByteSeparator-4    13.8ms ±20%   4.3ms ± 8%  -69.23%  (p=0.000 n=10+10)
SplitNSingleByteSeparator-4  1.88µs ± 9%  0.88µs ± 4%  -53.25%  (p=0.000 n=10+10)
SplitNMultiByteSeparator-4   4.83µs ±10%  1.32µs ± 9%  -72.65%  (p=0.000 n=10+10)

strings:

name                         old time/op  new time/op  delta
SplitSingleByteSeparator-4   21.4ms ± 8%   8.5ms ± 5%  -60.19%  (p=0.000 n=10+10)
SplitMultiByteSeparator-4    13.2ms ± 9%   3.9ms ± 4%  -70.29%  (p=0.000 n=10+10)
SplitNSingleByteSeparator-4  1.54µs ± 5%  0.75µs ± 7%  -51.21%  (p=0.000 n=10+10)
SplitNMultiByteSeparator-4   3.57µs ± 8%  1.01µs ±11%  -71.76%  (p=0.000 n=10+10)

Fixes #18973

Change-Id: Ie4bc010c6cc389001e72eab530497c81e5b26f34
Reviewed-on: https://go-review.googlesource.com/36510
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>

diff --git a/src/bytes/bytes.go b/src/bytes/bytes.go
index a273a919110c8fe9a6c5be280dd3042e37ceed56..029609afba2e777508e9b2be28bca1622e2de4b2 100644 (file)
--- a/src/bytes/bytes.go
+++ b/src/bytes/bytes.go
@@ -215,20 +215,21 @@ func genSplit(s, sep []byte, sepSave, n int) [][]byte {
        if n < 0 {
                n = Count(s, sep) + 1
        }
-       c := sep[0]
-       start := 0
+
        a := make([][]byte, n)
-       na := 0
-       for i := 0; i+len(sep) <= len(s) && na+1 < n; i++ {
-               if s[i] == c && (len(sep) == 1 || Equal(s[i:i+len(sep)], sep)) {
-                       a[na] = s[start : i+sepSave]
-                       na++
-                       start = i + len(sep)
-                       i += len(sep) - 1
+       n--
+       i := 0
+       for i < n {
+               m := Index(s, sep)
+               if m < 0 {
+                       break
                }
+               a[i] = s[:m+sepSave]
+               s = s[m+len(sep):]
+               i++
        }
-       a[na] = s[start:]
-       return a[0 : na+1]
+       a[i] = s
+       return a[:i+1]
 }
 
 // SplitN slices s into subslices separated by sep and returns a slice of

diff --git a/src/bytes/bytes_test.go b/src/bytes/bytes_test.go
index 26eac5e08c6df35cb6be2da6a84de310b2ff1ffc..dd8bdf2b04a203e834880df3ab05300f785ab1f4 100644 (file)
--- a/src/bytes/bytes_test.go
+++ b/src/bytes/bytes_test.go
@@ -1432,6 +1432,59 @@ func BenchmarkTrimSpace(b *testing.B) {
        }
 }
 
+func makeBenchInputHard() []byte {
+       tokens := [...]string{
+               "<a>", "<p>", "<b>", "<strong>",
+               "</a>", "</p>", "</b>", "</strong>",
+               "hello", "world",
+       }
+       x := make([]byte, 0, 1<<20)
+       for {
+               i := rand.Intn(len(tokens))
+               if len(x)+len(tokens[i]) >= 1<<20 {
+                       break
+               }
+               x = append(x, tokens[i]...)
+       }
+       return x
+}
+
+var benchInputHard = makeBenchInputHard()
+
+func BenchmarkSplitEmptySeparator(b *testing.B) {
+       for i := 0; i < b.N; i++ {
+               Split(benchInputHard, nil)
+       }
+}
+
+func BenchmarkSplitSingleByteSeparator(b *testing.B) {
+       sep := []byte("/")
+       for i := 0; i < b.N; i++ {
+               Split(benchInputHard, sep)
+       }
+}
+
+func BenchmarkSplitMultiByteSeparator(b *testing.B) {
+       sep := []byte("hello")
+       for i := 0; i < b.N; i++ {
+               Split(benchInputHard, sep)
+       }
+}
+
+func BenchmarkSplitNSingleByteSeparator(b *testing.B) {
+       sep := []byte("/")
+       for i := 0; i < b.N; i++ {
+               SplitN(benchInputHard, sep, 10)
+       }
+}
+
+func BenchmarkSplitNMultiByteSeparator(b *testing.B) {
+       sep := []byte("hello")
+       for i := 0; i < b.N; i++ {
+               SplitN(benchInputHard, sep, 10)
+       }
+}
+
 func BenchmarkRepeat(b *testing.B) {
        for i := 0; i < b.N; i++ {
                Repeat([]byte("-"), 80)

diff --git a/src/strings/strings.go b/src/strings/strings.go
index 2b1fbab5b2b8d2b2964d3a86b182e03687933b53..2165e15d8f57f079a8f0b15ec93d4ac42aeba094 100644 (file)
--- a/src/strings/strings.go
+++ b/src/strings/strings.go
@@ -239,20 +239,21 @@ func genSplit(s, sep string, sepSave, n int) []string {
        if n < 0 {
                n = Count(s, sep) + 1
        }
-       c := sep[0]
-       start := 0
+
        a := make([]string, n)
-       na := 0
-       for i := 0; i+len(sep) <= len(s) && na+1 < n; i++ {
-               if s[i] == c && (len(sep) == 1 || s[i:i+len(sep)] == sep) {
-                       a[na] = s[start : i+sepSave]
-                       na++
-                       start = i + len(sep)
-                       i += len(sep) - 1
-               }
+       n--
+       i := 0
+       for i < n {
+               m := Index(s, sep)
+               if m < 0 {
+                       break
+               }
+               a[i] = s[:m+sepSave]
+               s = s[m+len(sep):]
+               i++
        }
-       a[na] = s[start:]
-       return a[0 : na+1]
+       a[i] = s
+       return a[:i+1]
 }
 
 // SplitN slices s into substrings separated by sep and returns a slice of

diff --git a/src/strings/strings_test.go b/src/strings/strings_test.go
index 43979491c1fc58046719212a1388dd047f22bc40..3378d54fe20ac789ce4339363ab4b592926a8916 100644 (file)
--- a/src/strings/strings_test.go
+++ b/src/strings/strings_test.go
@@ -1476,24 +1476,36 @@ func BenchmarkFieldsFunc(b *testing.B) {
        }
 }
 
-func BenchmarkSplit1(b *testing.B) {
+func BenchmarkSplitEmptySeparator(b *testing.B) {
        for i := 0; i < b.N; i++ {
                Split(benchInputHard, "")
        }
 }
 
-func BenchmarkSplit2(b *testing.B) {
+func BenchmarkSplitSingleByteSeparator(b *testing.B) {
        for i := 0; i < b.N; i++ {
                Split(benchInputHard, "/")
        }
 }
 
-func BenchmarkSplit3(b *testing.B) {
+func BenchmarkSplitMultiByteSeparator(b *testing.B) {
        for i := 0; i < b.N; i++ {
                Split(benchInputHard, "hello")
        }
 }
 
+func BenchmarkSplitNSingleByteSeparator(b *testing.B) {
+       for i := 0; i < b.N; i++ {
+               SplitN(benchInputHard, "/", 10)
+       }
+}
+
+func BenchmarkSplitNMultiByteSeparator(b *testing.B) {
+       for i := 0; i < b.N; i++ {
+               SplitN(benchInputHard, "hello", 10)
+       }
+}
+
 func BenchmarkRepeat(b *testing.B) {
        for i := 0; i < b.N; i++ {
                Repeat("-", 80)