1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
|
// Copyright 2014 The Gogs Authors. All rights reserved.
// Use of this source code is governed by a MIT-style
// license that can be found in the LICENSE file.
package charset
import (
"bytes"
"fmt"
"unicode/utf8"
"code.gitea.io/gitea/modules/log"
"code.gitea.io/gitea/modules/setting"
"github.com/gogs/chardet"
"golang.org/x/net/html/charset"
"golang.org/x/text/transform"
)
// UTF8BOM is the utf-8 byte-order marker
var UTF8BOM = []byte{'\xef', '\xbb', '\xbf'}
// ToUTF8WithErr converts content to UTF8 encoding
func ToUTF8WithErr(content []byte) (string, error) {
charsetLabel, err := DetectEncoding(content)
if err != nil {
return "", err
} else if charsetLabel == "UTF-8" {
return string(RemoveBOMIfPresent(content)), nil
}
encoding, _ := charset.Lookup(charsetLabel)
if encoding == nil {
return string(content), fmt.Errorf("Unknown encoding: %s", charsetLabel)
}
// If there is an error, we concatenate the nicely decoded part and the
// original left over. This way we won't lose much data.
result, n, err := transform.Bytes(encoding.NewDecoder(), content)
if err != nil {
result = append(result, content[n:]...)
}
result = RemoveBOMIfPresent(result)
return string(result), err
}
// ToUTF8WithFallback detects the encoding of content and coverts to UTF-8 if possible
func ToUTF8WithFallback(content []byte) []byte {
charsetLabel, err := DetectEncoding(content)
if err != nil || charsetLabel == "UTF-8" {
return RemoveBOMIfPresent(content)
}
encoding, _ := charset.Lookup(charsetLabel)
if encoding == nil {
return content
}
// If there is an error, we concatenate the nicely decoded part and the
// original left over. This way we won't lose data.
result, n, err := transform.Bytes(encoding.NewDecoder(), content)
if err != nil {
return append(result, content[n:]...)
}
return RemoveBOMIfPresent(result)
}
// ToUTF8 converts content to UTF8 encoding and ignore error
func ToUTF8(content string) string {
res, _ := ToUTF8WithErr([]byte(content))
return res
}
// ToUTF8DropErrors makes sure the return string is valid utf-8; attempts conversion if possible
func ToUTF8DropErrors(content []byte) []byte {
charsetLabel, err := DetectEncoding(content)
if err != nil || charsetLabel == "UTF-8" {
return RemoveBOMIfPresent(content)
}
encoding, _ := charset.Lookup(charsetLabel)
if encoding == nil {
return content
}
// We ignore any non-decodable parts from the file.
// Some parts might be lost
var decoded []byte
decoder := encoding.NewDecoder()
idx := 0
for {
result, n, err := transform.Bytes(decoder, content[idx:])
decoded = append(decoded, result...)
if err == nil {
break
}
decoded = append(decoded, ' ')
idx = idx + n + 1
if idx >= len(content) {
break
}
}
return RemoveBOMIfPresent(decoded)
}
// RemoveBOMIfPresent removes a UTF-8 BOM from a []byte
func RemoveBOMIfPresent(content []byte) []byte {
if len(content) > 2 && bytes.Equal(content[0:3], UTF8BOM) {
return content[3:]
}
return content
}
// DetectEncoding detect the encoding of content
func DetectEncoding(content []byte) (string, error) {
if utf8.Valid(content) {
log.Debug("Detected encoding: utf-8 (fast)")
return "UTF-8", nil
}
textDetector := chardet.NewTextDetector()
var detectContent []byte
if len(content) < 1024 {
// Check if original content is valid
if _, err := textDetector.DetectBest(content); err != nil {
return "", err
}
times := 1024 / len(content)
detectContent = make([]byte, 0, times*len(content))
for i := 0; i < times; i++ {
detectContent = append(detectContent, content...)
}
} else {
detectContent = content
}
result, err := textDetector.DetectBest(detectContent)
if err != nil {
return "", err
}
// FIXME: to properly decouple this function the fallback ANSI charset should be passed as an argument
if result.Charset != "UTF-8" && len(setting.Repository.AnsiCharset) > 0 {
log.Debug("Using default AnsiCharset: %s", setting.Repository.AnsiCharset)
return setting.Repository.AnsiCharset, err
}
log.Debug("Detected encoding: %s", result.Charset)
return result.Charset, err
}
|
