repobase/repobase
1
0
Fork 0
Code Issues 8 Pull requests Projects 1 Releases Packages Wiki Activity Actions 4

#2282 fix utf-8 recognized as windows-1252

Browse source
This commit is contained in:
Unknwon 2015-12-25 05:25:47 -05:00
parent 13fe733037
commit 85af36332b
6 changed files with 13 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

4
modules/base/tool.go
View file

@ -25,6 +25,7 @@ import (
"golang.org/x/net/html/charset"
"github.com/gogits/gogs/modules/avatar"
"github.com/gogits/gogs/modules/log"
"github.com/gogits/gogs/modules/setting"
)
@ -52,7 +53,8 @@ func ShortSha(sha1 string) string {
}
func DetectEncoding(content []byte) string {
_, name, _ := charset.DetermineEncoding(content, setting.Repository.AnsiCharset)
_, name, certain := charset.DetermineEncoding(content, setting.Repository.AnsiCharset)
log.Debug("Detected encoding: %s (%v)", name, certain)
return name
}

10
modules/template/template.go
View file

@ -12,6 +12,7 @@ import (
"runtime"
"strings"
"time"
"unicode/utf8"
"golang.org/x/net/html/charset"
"golang.org/x/text/transform"
@ -130,20 +131,19 @@ func Sha1(str string) string {
}
func ToUtf8WithErr(content []byte) (error, string) {
charsetLabel := base.DetectEncoding(content)
if charsetLabel == "UTF-8" {
if utf8.Valid(content[:1024]) {
return nil, string(content)
}
charsetLabel := base.DetectEncoding(content)
encoding, _ := charset.Lookup(charsetLabel)
if encoding == nil {
return fmt.Errorf("unknown char decoder %s", charsetLabel), string(content)
return fmt.Errorf("Unknown encoding: %s", charsetLabel), string(content)
}
result, n, err := transform.String(encoding.NewDecoder(), string(content))
// If there is an error, we concatenate the nicely decoded part and the
// original left over. This way we won't loose data.
result, n, err := transform.String(encoding.NewDecoder(), string(content))
if err != nil {
result = result + string(content[n:])
}