351 lines
8.3 KiB
Go
351 lines
8.3 KiB
Go
// Copyright 2009 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 strconv
|
|
|
|
import (
|
|
"bytes"
|
|
"strings"
|
|
"unicode"
|
|
"unicode/utf8"
|
|
)
|
|
|
|
const lowerhex = "0123456789abcdef"
|
|
|
|
func quoteWith(s string, quote byte, ASCIIonly bool) string {
|
|
var buf bytes.Buffer
|
|
buf.WriteByte(quote)
|
|
for width := 0; len(s) > 0; s = s[width:] {
|
|
r := rune(s[0])
|
|
width = 1
|
|
if r >= utf8.RuneSelf {
|
|
r, width = utf8.DecodeRuneInString(s)
|
|
}
|
|
if width == 1 && r == utf8.RuneError {
|
|
buf.WriteString(`\x`)
|
|
buf.WriteByte(lowerhex[s[0]>>4])
|
|
buf.WriteByte(lowerhex[s[0]&0xF])
|
|
continue
|
|
}
|
|
if r == rune(quote) || r == '\\' { // always backslashed
|
|
buf.WriteByte('\\')
|
|
buf.WriteByte(byte(r))
|
|
continue
|
|
}
|
|
if ASCIIonly {
|
|
if r <= unicode.MaxASCII && unicode.IsPrint(r) {
|
|
buf.WriteRune(r)
|
|
continue
|
|
}
|
|
} else if unicode.IsPrint(r) {
|
|
buf.WriteRune(r)
|
|
continue
|
|
}
|
|
switch r {
|
|
case '\a':
|
|
buf.WriteString(`\a`)
|
|
case '\b':
|
|
buf.WriteString(`\b`)
|
|
case '\f':
|
|
buf.WriteString(`\f`)
|
|
case '\n':
|
|
buf.WriteString(`\n`)
|
|
case '\r':
|
|
buf.WriteString(`\r`)
|
|
case '\t':
|
|
buf.WriteString(`\t`)
|
|
case '\v':
|
|
buf.WriteString(`\v`)
|
|
default:
|
|
switch {
|
|
case r < ' ':
|
|
buf.WriteString(`\x`)
|
|
buf.WriteByte(lowerhex[s[0]>>4])
|
|
buf.WriteByte(lowerhex[s[0]&0xF])
|
|
case r > unicode.MaxRune:
|
|
r = 0xFFFD
|
|
fallthrough
|
|
case r < 0x10000:
|
|
buf.WriteString(`\u`)
|
|
for s := 12; s >= 0; s -= 4 {
|
|
buf.WriteByte(lowerhex[r>>uint(s)&0xF])
|
|
}
|
|
default:
|
|
buf.WriteString(`\U`)
|
|
for s := 28; s >= 0; s -= 4 {
|
|
buf.WriteByte(lowerhex[r>>uint(s)&0xF])
|
|
}
|
|
}
|
|
}
|
|
}
|
|
buf.WriteByte(quote)
|
|
return buf.String()
|
|
|
|
}
|
|
|
|
// Quote returns a double-quoted Go string literal representing s. The
|
|
// returned string uses Go escape sequences (\t, \n, \xFF, \u0100) for
|
|
// control characters and non-printable characters as defined by
|
|
// unicode.IsPrint.
|
|
func Quote(s string) string {
|
|
return quoteWith(s, '"', false)
|
|
}
|
|
|
|
// AppendQuote appends a double-quoted Go string literal representing s,
|
|
// as generated by Quote, to dst and returns the extended buffer.
|
|
func AppendQuote(dst []byte, s string) []byte {
|
|
return append(dst, Quote(s)...)
|
|
}
|
|
|
|
// QuoteToASCII returns a double-quoted Go string literal representing s.
|
|
// The returned string uses Go escape sequences (\t, \n, \xFF, \u0100) for
|
|
// non-ASCII characters and non-printable characters as defined by
|
|
// unicode.IsPrint.
|
|
func QuoteToASCII(s string) string {
|
|
return quoteWith(s, '"', true)
|
|
}
|
|
|
|
// AppendQuoteToASCII appends a double-quoted Go string literal representing s,
|
|
// as generated by QuoteToASCII, to dst and returns the extended buffer.
|
|
func AppendQuoteToASCII(dst []byte, s string) []byte {
|
|
return append(dst, QuoteToASCII(s)...)
|
|
}
|
|
|
|
// QuoteRune returns a single-quoted Go character literal representing the
|
|
// rune. The returned string uses Go escape sequences (\t, \n, \xFF, \u0100)
|
|
// for control characters and non-printable characters as defined by
|
|
// unicode.IsPrint.
|
|
func QuoteRune(r rune) string {
|
|
// TODO: avoid the allocation here.
|
|
return quoteWith(string(r), '\'', false)
|
|
}
|
|
|
|
// AppendQuoteRune appends a single-quoted Go character literal representing the rune,
|
|
// as generated by QuoteRune, to dst and returns the extended buffer.
|
|
func AppendQuoteRune(dst []byte, r rune) []byte {
|
|
return append(dst, QuoteRune(r)...)
|
|
}
|
|
|
|
// QuoteRuneToASCII returns a single-quoted Go character literal representing
|
|
// the rune. The returned string uses Go escape sequences (\t, \n, \xFF,
|
|
// \u0100) for non-ASCII characters and non-printable characters as defined
|
|
// by unicode.IsPrint.
|
|
func QuoteRuneToASCII(r rune) string {
|
|
// TODO: avoid the allocation here.
|
|
return quoteWith(string(r), '\'', true)
|
|
}
|
|
|
|
// AppendQuoteRune appends a single-quoted Go character literal representing the rune,
|
|
// as generated by QuoteRuneToASCII, to dst and returns the extended buffer.
|
|
func AppendQuoteRuneToASCII(dst []byte, r rune) []byte {
|
|
return append(dst, QuoteRuneToASCII(r)...)
|
|
}
|
|
|
|
// CanBackquote returns whether the string s would be
|
|
// a valid Go string literal if enclosed in backquotes.
|
|
func CanBackquote(s string) bool {
|
|
for i := 0; i < len(s); i++ {
|
|
if (s[i] < ' ' && s[i] != '\t') || s[i] == '`' {
|
|
return false
|
|
}
|
|
}
|
|
return true
|
|
}
|
|
|
|
func unhex(b byte) (v rune, ok 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
|
|
}
|
|
|
|
// UnquoteChar decodes the first character or byte in the escaped string
|
|
// or character literal represented by the string s.
|
|
// It returns four values:
|
|
//
|
|
// 1) value, the decoded Unicode code point or byte value;
|
|
// 2) multibyte, a boolean indicating whether the decoded character requires a multibyte UTF-8 representation;
|
|
// 3) tail, the remainder of the string after the character; and
|
|
// 4) an error that will be nil if the character is syntactically valid.
|
|
//
|
|
// The second argument, quote, specifies the type of literal being parsed
|
|
// and therefore which escaped quote character is permitted.
|
|
// If set to a single quote, it permits the sequence \' and disallows unescaped '.
|
|
// If set to a double quote, it permits \" and disallows unescaped ".
|
|
// If set to zero, it does not permit either escape and allows both quote characters to appear unescaped.
|
|
func UnquoteChar(s string, quote byte) (value rune, multibyte bool, tail string, err error) {
|
|
// easy cases
|
|
switch c := s[0]; {
|
|
case c == quote && (quote == '\'' || quote == '"'):
|
|
err = ErrSyntax
|
|
return
|
|
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
|
|
}
|
|
|
|
// hard case: c is backslash
|
|
if len(s) <= 1 {
|
|
err = ErrSyntax
|
|
return
|
|
}
|
|
c := s[1]
|
|
s = s[2:]
|
|
|
|
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 'x', 'u', 'U':
|
|
n := 0
|
|
switch c {
|
|
case 'x':
|
|
n = 2
|
|
case 'u':
|
|
n = 4
|
|
case 'U':
|
|
n = 8
|
|
}
|
|
var v rune
|
|
if len(s) < n {
|
|
err = ErrSyntax
|
|
return
|
|
}
|
|
for j := 0; j < n; j++ {
|
|
x, ok := unhex(s[j])
|
|
if !ok {
|
|
err = ErrSyntax
|
|
return
|
|
}
|
|
v = v<<4 | x
|
|
}
|
|
s = s[n:]
|
|
if c == 'x' {
|
|
// single-byte string, possibly not UTF-8
|
|
value = v
|
|
break
|
|
}
|
|
if v > unicode.MaxRune {
|
|
err = ErrSyntax
|
|
return
|
|
}
|
|
value = v
|
|
multibyte = true
|
|
case '0', '1', '2', '3', '4', '5', '6', '7':
|
|
v := rune(c) - '0'
|
|
if len(s) < 2 {
|
|
err = ErrSyntax
|
|
return
|
|
}
|
|
for j := 0; j < 2; j++ { // one digit already; two more
|
|
x := rune(s[j]) - '0'
|
|
if x < 0 || x > 7 {
|
|
err = ErrSyntax
|
|
return
|
|
}
|
|
v = (v << 3) | x
|
|
}
|
|
s = s[2:]
|
|
if v > 255 {
|
|
err = ErrSyntax
|
|
return
|
|
}
|
|
value = v
|
|
case '\\':
|
|
value = '\\'
|
|
case '\'', '"':
|
|
if c != quote {
|
|
err = ErrSyntax
|
|
return
|
|
}
|
|
value = rune(c)
|
|
default:
|
|
err = ErrSyntax
|
|
return
|
|
}
|
|
tail = s
|
|
return
|
|
}
|
|
|
|
// Unquote interprets s as a single-quoted, double-quoted,
|
|
// or backquoted Go string literal, returning the string value
|
|
// that s quotes. (If s is single-quoted, it would be a Go
|
|
// character literal; Unquote returns the corresponding
|
|
// one-character string.)
|
|
func Unquote(s string) (t string, err error) {
|
|
n := len(s)
|
|
if n < 2 {
|
|
return "", ErrSyntax
|
|
}
|
|
quote := s[0]
|
|
if quote != s[n-1] {
|
|
return "", ErrSyntax
|
|
}
|
|
s = s[1 : n-1]
|
|
|
|
if quote == '`' {
|
|
if strings.Contains(s, "`") {
|
|
return "", ErrSyntax
|
|
}
|
|
return s, nil
|
|
}
|
|
if quote != '"' && quote != '\'' {
|
|
return "", ErrSyntax
|
|
}
|
|
if strings.Index(s, "\n") >= 0 {
|
|
return "", ErrSyntax
|
|
}
|
|
|
|
// Is it trivial? Avoid allocation.
|
|
if strings.Index(s, `\`) < 0 && strings.IndexRune(s, rune(quote)) < 0 {
|
|
switch quote {
|
|
case '"':
|
|
return s, nil
|
|
case '\'':
|
|
r, size := utf8.DecodeRuneInString(s)
|
|
if size == len(s) && (r != utf8.RuneError || size != 1) {
|
|
return s, nil
|
|
}
|
|
}
|
|
}
|
|
|
|
var buf bytes.Buffer
|
|
for len(s) > 0 {
|
|
c, multibyte, ss, err := UnquoteChar(s, quote)
|
|
if err != nil {
|
|
return "", err
|
|
}
|
|
s = ss
|
|
if c < utf8.RuneSelf || !multibyte {
|
|
buf.WriteByte(byte(c))
|
|
} else {
|
|
buf.WriteString(string(c))
|
|
}
|
|
if quote == '\'' && len(s) != 0 {
|
|
// single-quoted must be single character
|
|
return "", ErrSyntax
|
|
}
|
|
}
|
|
return buf.String(), nil
|
|
}
|