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
|
package toml
import (
"unicode/utf8"
)
type utf8Err struct {
Index int
Size int
}
func (u utf8Err) Zero() bool {
return u.Size == 0
}
// Verified that a given string is only made of valid UTF-8 characters allowed
// by the TOML spec:
//
// Any Unicode character may be used except those that must be escaped:
// quotation mark, backslash, and the control characters other than tab (U+0000
// to U+0008, U+000A to U+001F, U+007F).
//
// It is a copy of the Go 1.17 utf8.Valid implementation, tweaked to exit early
// when a character is not allowed.
//
// The returned utf8Err is Zero() if the string is valid, or contains the byte
// index and size of the invalid character.
//
// quotation mark => already checked
// backslash => already checked
// 0-0x8 => invalid
// 0x9 => tab, ok
// 0xA - 0x1F => invalid
// 0x7F => invalid
func utf8TomlValidAlreadyEscaped(p []byte) (err utf8Err) {
// Fast path. Check for and skip 8 bytes of ASCII characters per iteration.
offset := 0
for len(p) >= 8 {
// Combining two 32 bit loads allows the same code to be used
// for 32 and 64 bit platforms.
// The compiler can generate a 32bit load for first32 and second32
// on many platforms. See test/codegen/memcombine.go.
first32 := uint32(p[0]) | uint32(p[1])<<8 | uint32(p[2])<<16 | uint32(p[3])<<24
second32 := uint32(p[4]) | uint32(p[5])<<8 | uint32(p[6])<<16 | uint32(p[7])<<24
if (first32|second32)&0x80808080 != 0 {
// Found a non ASCII byte (>= RuneSelf).
break
}
for i, b := range p[:8] {
if invalidAscii(b) {
err.Index = offset + i
err.Size = 1
return
}
}
p = p[8:]
offset += 8
}
n := len(p)
for i := 0; i < n; {
pi := p[i]
if pi < utf8.RuneSelf {
if invalidAscii(pi) {
err.Index = offset + i
err.Size = 1
return
}
i++
continue
}
x := first[pi]
if x == xx {
// Illegal starter byte.
err.Index = offset + i
err.Size = 1
return
}
size := int(x & 7)
if i+size > n {
// Short or invalid.
err.Index = offset + i
err.Size = n - i
return
}
accept := acceptRanges[x>>4]
if c := p[i+1]; c < accept.lo || accept.hi < c {
err.Index = offset + i
err.Size = 2
return
} else if size == 2 {
} else if c := p[i+2]; c < locb || hicb < c {
err.Index = offset + i
err.Size = 3
return
} else if size == 3 {
} else if c := p[i+3]; c < locb || hicb < c {
err.Index = offset + i
err.Size = 4
return
}
i += size
}
return
}
// Return the size of the next rune if valid, 0 otherwise.
func utf8ValidNext(p []byte) int {
c := p[0]
if c < utf8.RuneSelf {
if invalidAscii(c) {
return 0
}
return 1
}
x := first[c]
if x == xx {
// Illegal starter byte.
return 0
}
size := int(x & 7)
if size > len(p) {
// Short or invalid.
return 0
}
accept := acceptRanges[x>>4]
if c := p[1]; c < accept.lo || accept.hi < c {
return 0
} else if size == 2 {
} else if c := p[2]; c < locb || hicb < c {
return 0
} else if size == 3 {
} else if c := p[3]; c < locb || hicb < c {
return 0
}
return size
}
var invalidAsciiTable = [256]bool{
0x00: true,
0x01: true,
0x02: true,
0x03: true,
0x04: true,
0x05: true,
0x06: true,
0x07: true,
0x08: true,
// 0x09 TAB
// 0x0A LF
0x0B: true,
0x0C: true,
// 0x0D CR
0x0E: true,
0x0F: true,
0x10: true,
0x11: true,
0x12: true,
0x13: true,
0x14: true,
0x15: true,
0x16: true,
0x17: true,
0x18: true,
0x19: true,
0x1A: true,
0x1B: true,
0x1C: true,
0x1D: true,
0x1E: true,
0x1F: true,
// 0x20 - 0x7E Printable ASCII characters
0x7F: true,
}
func invalidAscii(b byte) bool {
return invalidAsciiTable[b]
}
// acceptRange gives the range of valid values for the second byte in a UTF-8
// sequence.
type acceptRange struct {
lo uint8 // lowest value for second byte.
hi uint8 // highest value for second byte.
}
// acceptRanges has size 16 to avoid bounds checks in the code that uses it.
var acceptRanges = [16]acceptRange{
0: {locb, hicb},
1: {0xA0, hicb},
2: {locb, 0x9F},
3: {0x90, hicb},
4: {locb, 0x8F},
}
// first is information about the first byte in a UTF-8 sequence.
var first = [256]uint8{
// 1 2 3 4 5 6 7 8 9 A B C D E F
as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, // 0x00-0x0F
as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, // 0x10-0x1F
as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, // 0x20-0x2F
as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, // 0x30-0x3F
as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, // 0x40-0x4F
as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, // 0x50-0x5F
as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, // 0x60-0x6F
as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, as, // 0x70-0x7F
// 1 2 3 4 5 6 7 8 9 A B C D E F
xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, // 0x80-0x8F
xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, // 0x90-0x9F
xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, // 0xA0-0xAF
xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, // 0xB0-0xBF
xx, xx, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, // 0xC0-0xCF
s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, s1, // 0xD0-0xDF
s2, s3, s3, s3, s3, s3, s3, s3, s3, s3, s3, s3, s3, s4, s3, s3, // 0xE0-0xEF
s5, s6, s6, s6, s7, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, xx, // 0xF0-0xFF
}
const (
// The default lowest and highest continuation byte.
locb = 0b10000000
hicb = 0b10111111
// These names of these constants are chosen to give nice alignment in the
// table below. The first nibble is an index into acceptRanges or F for
// special one-byte cases. The second nibble is the Rune length or the
// Status for the special one-byte case.
xx = 0xF1 // invalid: size 1
as = 0xF0 // ASCII: size 1
s1 = 0x02 // accept 0, size 2
s2 = 0x13 // accept 1, size 3
s3 = 0x03 // accept 0, size 3
s4 = 0x23 // accept 2, size 3
s5 = 0x34 // accept 3, size 4
s6 = 0x04 // accept 0, size 4
s7 = 0x44 // accept 4, size 4
)
|
