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
|
package lexer
import (
"fmt"
"strings"
"unicode"
"unicode/utf8"
)
func IsSpace(r rune) bool {
return unicode.IsSpace(r)
}
func IsAlphaNumeric(r rune) bool {
return IsAlphabetic(r) || unicode.IsDigit(r)
}
func IsAlphabetic(r rune) bool {
return r == '_' || r == '$' || unicode.IsLetter(r)
}
var (
newlineNormalizer = strings.NewReplacer("\r\n", "\n", "\r", "\n")
)
// Unescape takes a quoted string, unquotes, and unescapes it.
func unescape(value string) (string, error) {
// All strings normalize newlines to the \n representation.
value = newlineNormalizer.Replace(value)
n := len(value)
// Nothing to unescape / decode.
if n < 2 {
return value, fmt.Errorf("unable to unescape string")
}
// Quoted string of some form, must have same first and last char.
if value[0] != value[n-1] || (value[0] != '"' && value[0] != '\'') {
return value, fmt.Errorf("unable to unescape string")
}
value = value[1 : n-1]
// The string contains escape characters.
// The following logic is adapted from `strconv/quote.go`
var runeTmp [utf8.UTFMax]byte
buf := make([]byte, 0, 3*n/2)
for len(value) > 0 {
c, multibyte, rest, err := unescapeChar(value)
if err != nil {
return "", err
}
value = rest
if c < utf8.RuneSelf || !multibyte {
buf = append(buf, byte(c))
} else {
n := utf8.EncodeRune(runeTmp[:], c)
buf = append(buf, runeTmp[:n]...)
}
}
return string(buf), nil
}
// unescapeChar takes a string input and returns the following info:
//
// value - the escaped unicode rune at the front of the string.
// multibyte - whether the rune value might require multiple bytes to represent.
// tail - the remainder of the input string.
// err - error value, if the character could not be unescaped.
//
// When multibyte is true the return value may still fit within a single byte,
// but a multibyte conversion is attempted which is more expensive than when the
// value is known to fit within one byte.
func unescapeChar(s string) (value rune, multibyte bool, tail string, err error) {
// 1. Character is not an escape sequence.
switch c := s[0]; {
case c >= utf8.RuneSelf:
r, size := utf8.DecodeRuneInString(s)
return r, true, s[size:], nil
case c != '\\':
return rune(s[0]), false, s[1:], nil
}
// 2. Last character is the start of an escape sequence.
if len(s) <= 1 {
err = fmt.Errorf("unable to unescape string, found '\\' as last character")
return
}
c := s[1]
s = s[2:]
// 3. Common escape sequences shared with Google SQL
switch c {
case 'a':
value = '\a'
case 'b':
value = '\b'
case 'f':
value = '\f'
case 'n':
value = '\n'
case 'r':
value = '\r'
case 't':
value = '\t'
case 'v':
value = '\v'
case '\\':
value = '\\'
case '\'':
value = '\''
case '"':
value = '"'
case '`':
value = '`'
case '?':
value = '?'
// 4. Unicode escape sequences, reproduced from `strconv/quote.go`
case 'x', 'X', 'u', 'U':
n := 0
switch c {
case 'x', 'X':
n = 2
case 'u':
n = 4
case 'U':
n = 8
}
var v rune
if len(s) < n {
err = fmt.Errorf("unable to unescape string")
return
}
for j := 0; j < n; j++ {
x, ok := unhex(s[j])
if !ok {
err = fmt.Errorf("unable to unescape string")
return
}
v = v<<4 | x
}
s = s[n:]
if v > utf8.MaxRune {
err = fmt.Errorf("unable to unescape string")
return
}
value = v
multibyte = true
// 5. Octal escape sequences, must be three digits \[0-3][0-7][0-7]
case '0', '1', '2', '3':
if len(s) < 2 {
err = fmt.Errorf("unable to unescape octal sequence in string")
return
}
v := rune(c - '0')
for j := 0; j < 2; j++ {
x := s[j]
if x < '0' || x > '7' {
err = fmt.Errorf("unable to unescape octal sequence in string")
return
}
v = v*8 + rune(x-'0')
}
if v > utf8.MaxRune {
err = fmt.Errorf("unable to unescape string")
return
}
value = v
s = s[2:]
multibyte = true
// Unknown escape sequence.
default:
err = fmt.Errorf("unable to unescape string")
}
tail = s
return
}
func unhex(b byte) (rune, bool) {
c := rune(b)
switch {
case '0' <= c && c <= '9':
return c - '0', true
case 'a' <= c && c <= 'f':
return c - 'a' + 10, true
case 'A' <= c && c <= 'F':
return c - 'A' + 10, true
}
return 0, false
}
|
