Simplify escapeHTML and uncomment it's benchmark

Simplify and optimize escapeHTML as per @Ambrevar's suggestion: lean on
the fact that we're dealing with bytes: declare a 256-element array with
most of it filled with zeros, except for the few slots that need
escaping. This avoids some conditionals in a tight loop.

Also, uncomment it's benchmark.
This commit is contained in:
Vytautas Šaltenis 2016-10-08 17:58:30 +03:00
parent d9ffdb7464
commit 461803619b
2 changed files with 11 additions and 24 deletions

27
esc.go
View File

@ -5,32 +5,21 @@ import (
"io"
)
type escMap struct {
char byte
seq []byte
}
var htmlEscaper = []escMap{
{'&', []byte("&")},
{'<', []byte("&lt;")},
{'>', []byte("&gt;")},
{'"', []byte("&quot;")},
var htmlEscaper = [256][]byte{
'&': []byte("&amp;"),
'<': []byte("&lt;"),
'>': []byte("&gt;"),
'"': []byte("&quot;"),
}
func escapeHTML(w io.Writer, s []byte) {
var start, end int
var sEnd byte
for end < len(s) {
sEnd = s[end]
if sEnd == '&' || sEnd == '<' || sEnd == '>' || sEnd == '"' {
for i := 0; i < len(htmlEscaper); i++ {
if sEnd == htmlEscaper[i].char {
escSeq := htmlEscaper[s[end]]
if escSeq != nil {
w.Write(s[start:end])
w.Write(htmlEscaper[i].seq)
w.Write(escSeq)
start = end + 1
break
}
}
}
end++
}

View File

@ -23,7 +23,6 @@ func TestEsc(t *testing.T) {
}
}
/*
func BenchmarkEscapeHTML(b *testing.B) {
tests := [][]byte{
[]byte(""),
@ -33,7 +32,7 @@ func BenchmarkEscapeHTML(b *testing.B) {
[]byte("4 < 5."),
[]byte("6 > 5."),
[]byte("Here's a [link] [1] with an ampersand in the URL."),
[]byte("Here's a link with an amersand in the link text: [AT&T] [2]."),
[]byte("Here's a link with an ampersand in the link text: [AT&T] [2]."),
[]byte("Here's an inline [link](/script?foo=1&bar=2)."),
[]byte("Here's an inline [link](</script?foo=1&bar=2>)."),
[]byte("[1]: http://example.com/?foo=1&bar=2"),
@ -47,4 +46,3 @@ func BenchmarkEscapeHTML(b *testing.B) {
}
}
}
*/