summaryrefslogtreecommitdiff
path: root/vendor/golang.org/x/net/html/charset/charset.go
blob: 13bed1599f71047c2542d156898c20fc6f1f51f0 (plain)
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
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
// Copyright 2013 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.

// Package charset provides common text encodings for HTML documents.
//
// The mapping from encoding labels to encodings is defined at
// https://encoding.spec.whatwg.org/.
package charset // import "golang.org/x/net/html/charset"

import (
	"bytes"
	"fmt"
	"io"
	"mime"
	"strings"
	"unicode/utf8"

	"golang.org/x/net/html"
	"golang.org/x/text/encoding"
	"golang.org/x/text/encoding/charmap"
	"golang.org/x/text/encoding/htmlindex"
	"golang.org/x/text/transform"
)

// Lookup returns the encoding with the specified label, and its canonical
// name. It returns nil and the empty string if label is not one of the
// standard encodings for HTML. Matching is case-insensitive and ignores
// leading and trailing whitespace. Encoders will use HTML escape sequences for
// runes that are not supported by the character set.
func Lookup(label string) (e encoding.Encoding, name string) {
	e, err := htmlindex.Get(label)
	if err != nil {
		return nil, ""
	}
	name, _ = htmlindex.Name(e)
	return &htmlEncoding{e}, name
}

type htmlEncoding struct{ encoding.Encoding }

func (h *htmlEncoding) NewEncoder() *encoding.Encoder {
	// HTML requires a non-terminating legacy encoder. We use HTML escapes to
	// substitute unsupported code points.
	return encoding.HTMLEscapeUnsupported(h.Encoding.NewEncoder())
}

// DetermineEncoding determines the encoding of an HTML document by examining
// up to the first 1024 bytes of content and the declared Content-Type.
//
// See http://www.whatwg.org/specs/web-apps/current-work/multipage/parsing.html#determining-the-character-encoding
func DetermineEncoding(content []byte, contentType string) (e encoding.Encoding, name string, certain bool) {
	if len(content) > 1024 {
		content = content[:1024]
	}

	for _, b := range boms {
		if bytes.HasPrefix(content, b.bom) {
			e, name = Lookup(b.enc)
			return e, name, true
		}
	}

	if _, params, err := mime.ParseMediaType(contentType); err == nil {
		if cs, ok := params["charset"]; ok {
			if e, name = Lookup(cs); e != nil {
				return e, name, true
			}
		}
	}

	if len(content) > 0 {
		e, name = prescan(content)
		if e != nil {
			return e, name, false
		}
	}

	// Try to detect UTF-8.
	// First eliminate any partial rune at the end.
	for i := len(content) - 1; i >= 0 && i > len(content)-4; i-- {
		b := content[i]
		if b < 0x80 {
			break
		}
		if utf8.RuneStart(b) {
			content = content[:i]
			break
		}
	}
	hasHighBit := false
	for _, c := range content {
		if c >= 0x80 {
			hasHighBit = true
			break
		}
	}
	if hasHighBit && utf8.Valid(content) {
		return encoding.Nop, "utf-8", false
	}

	// TODO: change default depending on user's locale?
	return charmap.Windows1252, "windows-1252", false
}

// NewReader returns an io.Reader that converts the content of r to UTF-8.
// It calls DetermineEncoding to find out what r's encoding is.
func NewReader(r io.Reader, contentType string) (io.Reader, error) {
	preview := make([]byte, 1024)
	n, err := io.ReadFull(r, preview)
	switch {
	case err == io.ErrUnexpectedEOF:
		preview = preview[:n]
		r = bytes.NewReader(preview)
	case err != nil:
		return nil, err
	default:
		r = io.MultiReader(bytes.NewReader(preview), r)
	}

	if e, _, _ := DetermineEncoding(preview, contentType); e != encoding.Nop {
		r = transform.NewReader(r, e.NewDecoder())
	}
	return r, nil
}

// NewReaderLabel returns a reader that converts from the specified charset to
// UTF-8. It uses Lookup to find the encoding that corresponds to label, and
// returns an error if Lookup returns nil. It is suitable for use as
// encoding/xml.Decoder's CharsetReader function.
func NewReaderLabel(label string, input io.Reader) (io.Reader, error) {
	e, _ := Lookup(label)
	if e == nil {
		return nil, fmt.Errorf("unsupported charset: %q", label)
	}
	return transform.NewReader(input, e.NewDecoder()), nil
}

func prescan(content []byte) (e encoding.Encoding, name string) {
	z := html.NewTokenizer(bytes.NewReader(content))
	for {
		switch z.Next() {
		case html.ErrorToken:
			return nil, ""

		case html.StartTagToken, html.SelfClosingTagToken:
			tagName, hasAttr := z.TagName()
			if !bytes.Equal(tagName, []byte("meta")) {
				continue
			}
			attrList := make(map[string]bool)
			gotPragma := false

			const (
				dontKnow = iota
				doNeedPragma
				doNotNeedPragma
			)
			needPragma := dontKnow

			name = ""
			e = nil
			for hasAttr {
				var key, val []byte
				key, val, hasAttr = z.TagAttr()
				ks := string(key)
				if attrList[ks] {
					continue
				}
				attrList[ks] = true
				for i, c := range val {
					if 'A' <= c && c <= 'Z' {
						val[i] = c + 0x20
					}
				}

				switch ks {
				case "http-equiv":
					if bytes.Equal(val, []byte("content-type")) {
						gotPragma = true
					}

				case "content":
					if e == nil {
						name = fromMetaElement(string(val))
						if name != "" {
							e, name = Lookup(name)
							if e != nil {
								needPragma = doNeedPragma
							}
						}
					}

				case "charset":
					e, name = Lookup(string(val))
					needPragma = doNotNeedPragma
				}
			}

			if needPragma == dontKnow || needPragma == doNeedPragma && !gotPragma {
				continue
			}

			if strings.HasPrefix(name, "utf-16") {
				name = "utf-8"
				e = encoding.Nop
			}

			if e != nil {
				return e, name
			}
		}
	}
}

func fromMetaElement(s string) string {
	for s != "" {
		csLoc := strings.Index(s, "charset")
		if csLoc == -1 {
			return ""
		}
		s = s[csLoc+len("charset"):]
		s = strings.TrimLeft(s, " \t\n\f\r")
		if !strings.HasPrefix(s, "=") {
			continue
		}
		s = s[1:]
		s = strings.TrimLeft(s, " \t\n\f\r")
		if s == "" {
			return ""
		}
		if q := s[0]; q == '"' || q == '\'' {
			s = s[1:]
			closeQuote := strings.IndexRune(s, rune(q))
			if closeQuote == -1 {
				return ""
			}
			return s[:closeQuote]
		}

		end := strings.IndexAny(s, "; \t\n\f\r")
		if end == -1 {
			end = len(s)
		}
		return s[:end]
	}
	return ""
}

var boms = []struct {
	bom []byte
	enc string
}{
	{[]byte{0xfe, 0xff}, "utf-16be"},
	{[]byte{0xff, 0xfe}, "utf-16le"},
	{[]byte{0xef, 0xbb, 0xbf}, "utf-8"},
}