From: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Date: Fri, 13 Sep 2024 03:15:51 +0000 (+0700)
Subject: unicode/utf8: speedup RuneCount
X-Git-Tag: go1.24rc1~898
X-Git-Url: http://www.git.cypherpunks.su/?a=commitdiff_plain;h=3d33437c450aa74014ea1d41cd986b6ee6266984;p=gostls13.git

unicode/utf8: speedup RuneCount

CL 612617 did speedup RuneCountInString, thus we can now use it to
speedup RuneCount, too.

name                         old time/op    new time/op    delta
RuneCountTenASCIIChars-8       8.69ns ± 1%    3.59ns ± 2%  -58.66%  (p=0.000 n=9+9)
RuneCountTenJapaneseChars-8    49.8ns ± 2%    40.9ns ± 0%  -17.94%  (p=0.000 n=10+8)

Change-Id: I311750c00efc79af35fb0ca3b482a5d94e0a7977
Reviewed-on: https://go-review.googlesource.com/c/go/+/612955
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Tim King <taking@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
---

diff --git a/src/unicode/utf8/utf8.go b/src/unicode/utf8/utf8.go
index 9743b74258..180c008ed5 100644
--- a/src/unicode/utf8/utf8.go
+++ b/src/unicode/utf8/utf8.go
@@ -414,35 +414,11 @@ func appendRuneNonASCII(p []byte, r rune) []byte {
 func RuneCount(p []byte) int {
 	np := len(p)
 	var n int
-	for i := 0; i < np; {
-		n++
-		c := p[i]
-		if c < RuneSelf {
-			// ASCII fast path
-			i++
-			continue
-		}
-		x := first[c]
-		if x == xx {
-			i++ // invalid.
-			continue
-		}
-		size := int(x & 7)
-		if i+size > np {
-			i++ // Short or invalid.
-			continue
+	for ; n < np; n++ {
+		if c := p[n]; c >= RuneSelf {
+			// non-ASCII slow path
+			return n + RuneCountInString(string(p[n:]))
 		}
-		accept := acceptRanges[x>>4]
-		if c := p[i+1]; c < accept.lo || accept.hi < c {
-			size = 1
-		} else if size == 2 {
-		} else if c := p[i+2]; c < locb || hicb < c {
-			size = 1
-		} else if size == 3 {
-		} else if c := p[i+3]; c < locb || hicb < c {
-			size = 1
-		}
-		i += size
 	}
 	return n
 }