gofront 1.8.0 → 1.9.0

package/stdlib/strconv/quote.go

@@ -4,45 +4,95 @@
 
 package strconv
 
-import (
-	"unicode"
-	"unicode/utf8"
-)
-
-// IntSize is the size in bits of an int or uint value.
-const IntSize = 64
+import "unicode/utf8"
 
 const lowerhex = "0123456789abcdef"
 
-// A NumError records a failed conversion.
-type NumError struct {
-	Func string // the failing function (ParseBool, ParseInt, ParseUint, ParseFloat, ParseComplex)
-	Num  string // the input
-	Err  error  // the reason the conversion failed (e.g. ErrRange, ErrSyntax, etc.)
+func bsearch16(s []uint16, v uint16) (int, bool) {
+	n := len(s)
+	i, j := 0, n
+	for i < j {
+		h := i + (j-i)>>1
+		if s[h] < v {
+			i = h + 1
+		} else {
+			j = h
+		}
+	}
+	return i, i < n && s[i] == v
 }
 
-func (e *NumError) Error() string {
-	return "strconv." + e.Func + ": parsing " + Quote(e.Num) + ": " + e.Err.Error()
+func bsearch32(s []uint32, v uint32) (int, bool) {
+	n := len(s)
+	i, j := 0, n
+	for i < j {
+		h := i + (j-i)>>1
+		if s[h] < v {
+			i = h + 1
+		} else {
+			j = h
+		}
+	}
+	return i, i < n && s[i] == v
 }
 
-func (e *NumError) Unwrap() error { return e.Err }
-
 // IsPrint reports whether the rune is defined as printable by Go, with
-// the same definition as unicode.IsPrint: letters, numbers, punctuation,
+// the same definition as [unicode.IsPrint]: letters, numbers, punctuation,
 // symbols and ASCII space.
 func IsPrint(r rune) bool {
-	if 0x20 <= r && r <= 0x7E {
+	// Fast check for Latin-1
+	if r <= 0xFF {
+		if 0x20 <= r && r <= 0x7E {
+			// All the ASCII is printable from space through DEL-1.
+			return true
+		}
+		if 0xA1 <= r && r <= 0xFF {
+			// Similarly for ¡ through ÿ...
+			return r != 0xAD // ...except for the bizarre soft hyphen.
+		}
+		return false
+	}
+
+	// First, find first i such that isPrint[i] >= x.
+	// This is the index of either the start or end of a pair that might span x.
+	// The start is even (isPrint[i&^1]) and the end is odd (isPrint[i|1]).
+	// If we find x in a range, make sure x is not in isNotPrint list.
+	if 0 <= r && r < 1<<16 {
+		rr := uint16(r)
+		i, _ := bsearch16(isPrint16, rr)
+		if i >= len(isPrint16) || rr < isPrint16[i&^1] || isPrint16[i|1] < rr {
+			return false
+		}
+		_, found := bsearch16(isNotPrint16, rr)
+		return !found
+	}
+
+	rr := uint32(r)
+	i, _ := bsearch32(isPrint32, rr)
+	if i >= len(isPrint32) || rr < isPrint32[i&^1] || isPrint32[i|1] < rr {
+		return false
+	}
+	if r >= 0x20000 {
 		return true
 	}
-	return unicode.IsPrint(r)
+	r -= 0x10000
+	_, found := bsearch16(isNotPrint32, uint16(r))
+	return !found
 }
 
-// IsGraphic reports whether the rune is defined as a Graphic by Unicode.
+// IsGraphic reports whether the rune is defined as a Graphic by Unicode. Such
+// characters include letters, marks, numbers, punctuation, symbols, and
+// spaces, from categories L, M, N, P, S, and Zs.
 func IsGraphic(r rune) bool {
-	if 0x20 <= r && r <= 0x7E {
+	if IsPrint(r) {
 		return true
 	}
-	return unicode.IsGraphic(r)
+	// We know r must fit in 16 bits - see makeisprint.go.
+	if r > 0xFFFF {
+		return false
+	}
+	_, found := bsearch16(isGraphic, uint16(r))
+	return found
 }
 
 func quoteWith(s string, quote byte, ASCIIonly, graphicOnly bool) string {
@@ -61,7 +111,9 @@ func appendQuotedWith(buf []byte, s string, quote byte, ASCIIonly, graphicOnly b
 	buf = append(buf, quote)
 	for len(s) > 0 {
 		r, width := utf8.DecodeRuneInString(s)
-		if width == 1 && r == utf8.RuneError {
+		// A U+FFFD that is in the string, not an invalid byte, has width 1 where
+		// a string is made of UTF-16 units.
+		if width == 1 && r == utf8.RuneError && rune(s[0]) != utf8.RuneError {
 			b0 := int(byte(s[0]))
 			buf = append(buf, byte('\\'), byte('x'), hexDigit(b0>>4), hexDigit(b0))
 			s = s[1:]
@@ -138,6 +190,13 @@ func appendEscapedRune(buf []byte, r rune, quote byte, ASCIIonly, graphicOnly bo
 	return buf
 }
 
+// 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 [IsPrint].
+func Quote(s string) string {
+	return quoteWith(s, byte('"'), false, false)
+}
+
 // QuoteRune returns a single-quoted Go character literal representing the rune.
 func QuoteRune(r rune) string {
 	return quoteRuneWith(r, byte('\''), false, false)
@@ -178,10 +237,10 @@ func CanBackquote(s string) bool {
 	for len(s) > 0 {
 		r, wid := utf8.DecodeRuneInString(s)
 		s = s[wid:]
+		if r == '\ufeff' {
+			return false
+		}
 		if wid > 1 {
-			if r == '\ufeff' {
-				return false
-			}
 			continue
 		}
 		if r == utf8.RuneError {
@@ -221,6 +280,9 @@ func unquoteHex(c byte, s string) (rune, bool, string, error) {
 	}
 	var v rune
 	for j := 0; j < n; j++ {
+		if s[j] >= utf8.RuneSelf {
+			return 0, false, "", ErrSyntax
+		}
 		x, ok := unhex(byte(s[j]))
 		if !ok {
 			return 0, false, "", ErrSyntax
@@ -243,6 +305,9 @@ func unquoteOctal(c byte, s string) (rune, string, error) {
 		return 0, "", ErrSyntax
 	}
 	for j := 0; j < 2; j++ {
+		if s[j] >= utf8.RuneSelf {
+			return 0, "", ErrSyntax
+		}
 		b := byte(s[j])
 		if b < byte('0') || b > byte('7') {
 			return 0, "", ErrSyntax
@@ -260,20 +325,24 @@ func UnquoteChar(s string, quote byte) (value rune, multibyte bool, tail string,
 	if len(s) == 0 {
 		return 0, false, "", ErrSyntax
 	}
+	// A string index is a UTF-16 unit: decide on it before narrowing to a byte.
+	if s[0] >= utf8.RuneSelf {
+		r, size := utf8.DecodeRuneInString(s)
+		return r, true, s[size:], nil
+	}
 	c := byte(s[0])
 	if c == quote && (quote == byte('\'') || quote == byte('"')) {
 		return 0, false, "", ErrSyntax
 	}
-	if rune(c) >= utf8.RuneSelf {
-		r, size := utf8.DecodeRuneInString(s)
-		return r, true, s[size:], nil
-	}
 	if c != byte('\\') {
 		return rune(c), false, s[1:], nil
 	}
 	if len(s) <= 1 {
 		return 0, false, "", ErrSyntax
 	}
+	if s[1] >= utf8.RuneSelf {
+		return 0, false, "", ErrSyntax
+	}
 	esc := byte(s[1])
 	s = s[2:]
 	switch esc {
@@ -309,33 +378,149 @@ func UnquoteChar(s string, quote byte) (value rune, multibyte bool, tail string,
 	}
 }
 
-// QuotedPrefix returns the quoted string at the prefix of s.
+// QuotedPrefix returns the quoted string (as understood by [Unquote]) at the prefix of s.
+// If s does not start with a valid quoted string, QuotedPrefix returns an error.
 func QuotedPrefix(s string) (string, error) {
-	if len(s) < 2 {
+	out, _, err := unquote(s, false)
+	return out, err
+}
+
+// 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. For an empty character literal
+// Unquote returns the empty string.)
+func Unquote(s string) (string, error) {
+	out, rem, err := unquote(s, true)
+	if len(rem) > 0 {
 		return "", ErrSyntax
 	}
-	quote := byte(s[0])
-	if quote != byte('"') && quote != byte('\'') && quote != byte('`') {
-		return "", ErrSyntax
+	return out, err
+}
+
+// unquote parses a quoted string at the start of the input,
+// returning the parsed prefix, the remaining suffix, and any parse errors.
+// If unescape is true, the parsed prefix is unescaped,
+// otherwise the input prefix is provided verbatim.
+func unquote(in string, unescape bool) (out, rem string, err error) {
+	// Determine the quote form and optimistically find the terminating quote.
+	if len(in) < 2 {
+		return "", in, ErrSyntax
+	}
+	if in[0] > 0x7F {
+		return "", in, ErrSyntax
 	}
-	if quote == byte('`') {
-		for i := 1; i < len(s); i++ {
-			if byte(s[i]) == byte('`') {
-				return s[:i+1], nil
+	quote := byte(in[0])
+	end := indexByte(in[1:], quote)
+	if end < 0 {
+		return "", in, ErrSyntax
+	}
+	end += 2 // position after terminating quote; may be wrong if escape sequences are present
+
+	switch quote {
+	case '`':
+		switch {
+		case !unescape:
+			out = in[:end] // include quotes
+		case !containsByte(in[:end], '\r'):
+			out = in[len("`") : end-len("`")] // exclude quotes
+		default:
+			// Carriage return characters ('\r') inside raw string literals
+			// are discarded from the raw string value.
+			// Built as a string: GoFront strings are indexed by UTF-16 unit, so
+			// a byte buffer would split multi-byte text.
+			out = ""
+			start := len("`")
+			for i := len("`"); i < end-len("`"); i++ {
+				if in[i] == '\r' {
+					out += in[start:i]
+					start = i + 1
+				}
 			}
+			out += in[start : end-len("`")]
 		}
-		return "", ErrSyntax
-	}
-	rem := s[1:]
-	for len(rem) > 0 {
-		if byte(rem[0]) == quote {
-			return s[:len(s)-len(rem)+1], nil
+		// NOTE: Prior implementations did not verify that raw strings consist
+		// of valid UTF-8 characters and we continue to not verify it as such.
+		// The Go specification does not explicitly require valid UTF-8,
+		// but only mention that it is implicitly valid for Go source code
+		// (which must be valid UTF-8).
+		return out, in[end:], nil
+	case '"', '\'':
+		// Handle quoted strings without any escape sequences.
+		if !containsByte(in[:end], '\\') && !containsByte(in[:end], '\n') {
+			var valid bool
+			switch quote {
+			case '"':
+				valid = utf8.ValidString(in[len(`"`) : end-len(`"`)])
+			case '\'':
+				r, n := utf8.DecodeRuneInString(in[len("'") : end-len("'")])
+				valid = len("'")+n+len("'") == end && (r != utf8.RuneError || n != 1)
+			}
+			if valid {
+				out = in[:end]
+				if unescape {
+					out = out[1 : end-1] // exclude quotes
+				}
+				return out, in[end:], nil
+			}
 		}
-		_, _, next, err := UnquoteChar(rem, quote)
-		if err != nil {
-			return "", err
+
+		// Handle quoted strings with escape sequences.
+		var buf []byte
+		in0 := in
+		in = in[1:] // skip starting quote
+		if unescape {
+			buf = make([]byte, 0, 3*end/2) // try to avoid more allocations
+		}
+		for len(in) > 0 && in[0] != quote {
+			// Process the next character,
+			// rejecting any unescaped newline characters which are invalid.
+			r, multibyte, rem, err := UnquoteChar(in, quote)
+			if in[0] == '\n' || err != nil {
+				return "", in0, ErrSyntax
+			}
+			in = rem
+
+			// Append the character if unescaping the input.
+			if unescape {
+				if r < utf8.RuneSelf || !multibyte {
+					buf = append(buf, byte(r))
+				} else {
+					buf = utf8.AppendRune(buf, r)
+				}
+			}
+
+			// Single quoted strings must be a single character.
+			if quote == '\'' {
+				break
+			}
+		}
+
+		// Verify that the string ends with a terminating quote.
+		if !(len(in) > 0 && in[0] == quote) {
+			return "", in0, ErrSyntax
+		}
+		in = in[1:] // skip terminating quote
+
+		if unescape {
+			return string(buf), in, nil
+		}
+		return in0[:len(in0)-len(in)], in, nil
+	default:
+		return "", in, ErrSyntax
+	}
+}
+
+func indexByte(s string, c byte) int {
+	for i := 0; i < len(s); i++ {
+		if s[i] == c {
+			return i
 		}
-		rem = next
 	}
-	return "", ErrSyntax
+	return -1
+}
+
+func containsByte(s string, c byte) bool {
+	return indexByte(s, c) >= 0
 }