qbit/go
1
0
Fork 0
mirror of https://github.com/golang/go synced 2024-11-13 12:20:26 -07:00
Code Releases Activity

html: use sync.OnceValues instead of var once sync.Once

Browse Source 
Simplify populateMaps with sync.OnceValues.

Change-Id: Id52e6e1623c621b8d51e11fecbe3f1fab1e74eb4
GitHub-Last-Rev: 3cf736ae29
GitHub-Pull-Request: golang/go#69946
Reviewed-on: https://go-review.googlesource.com/c/go/+/621255
Reviewed-by: Michael Pratt <mpratt@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Commit-Queue: Ian Lance Taylor <iant@google.com>
This commit is contained in:
1911860538 2024-10-21 16:27:03 +00:00 committed by Gopher Robot
parent 2e9ed44d39
commit 205ab8a3fe
3 changed files with 14 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
src/html/entity.go
View File

@ -9,22 +9,16 @@ import "sync"
// All entities that do not end with ';' are 6 or fewer bytes long.
const longestEntityWithoutSemicolon = 6
// entityMaps returns entity and entity2.
//
// entity is a map from HTML entity names to their values. The semicolon matters:
// https://html.spec.whatwg.org/multipage/named-characters.html
// lists both "amp" and "amp;" as two separate entries.
//
// Note that the HTML5 list is larger than the HTML4 list at
// http://www.w3.org/TR/html4/sgml/entities.html
var entity map[string]rune
// HTML entities that are two unicode codepoints.
var entity2 map[string][2]rune
// populateMapsOnce guards calling populateMaps.
var populateMapsOnce sync.Once
// populateMaps populates entity and entity2.
func populateMaps() {
//
// entity2 is a map of HTML entities to two unicode codepoints.
var entityMaps = sync.OnceValues(func() (entity map[string]rune, entity2 map[string][2]rune) {
entity = map[string]rune{
"AElig;": '\U000000C6',
"AMP;": '\U00000026',
@ -2262,4 +2256,6 @@ func populateMaps() {
"vsupnE;": {'\u2ACC', '\uFE00'},
"vsupne;": {'\u228B', '\uFE00'},
}
}
return entity, entity2
})

6
src/html/entity_test.go
View File

@ -9,11 +9,9 @@ import (
"unicode/utf8"
)
func init() {
UnescapeString("") // force load of entity maps
}
func TestEntityLength(t *testing.T) {
entity, entity2 := entityMaps()
if len(entity) == 0 || len(entity2) == 0 {
t.Fatal("maps not loaded")
}

8
src/html/escape.go
View File

@ -53,7 +53,7 @@ var replacementTable = [...]rune{
// unescapeEntity reads an entity like "&lt;" from b[src:] and writes the
// corresponding "<" to b[dst:], returning the incremented dst and src cursors.
// Precondition: b[src] == '&' && dst <= src.
func unescapeEntity(b []byte, dst, src int) (dst1, src1 int) {
func unescapeEntity(b []byte, dst, src int, entity map[string]rune, entity2 map[string][2]rune) (dst1, src1 int) {
const attribute = false
// http://www.whatwg.org/specs/web-apps/current-work/multipage/tokenization.html#consume-a-character-reference
@ -185,7 +185,6 @@ func EscapeString(s string) string {
// UnescapeString(EscapeString(s)) == s always holds, but the converse isn't
// always true.
func UnescapeString(s string) string {
populateMapsOnce.Do(populateMaps)
i := strings.IndexByte(s, '&')
if i < 0 {
@ -193,7 +192,8 @@ func UnescapeString(s string) string {
}
b := []byte(s)
dst, src := unescapeEntity(b, i, i)
entity, entity2 := entityMaps()
dst, src := unescapeEntity(b, i, i, entity, entity2)
for len(s[src:]) > 0 {
if s[src] == '&' {
i = 0
@ -208,7 +208,7 @@ func UnescapeString(s string) string {
if i > 0 {
copy(b[dst:], s[src:src+i])
}
dst, src = unescapeEntity(b, dst+i, src+i)
dst, src = unescapeEntity(b, dst+i, src+i, entity, entity2)
}
return string(b[:dst])
}