Add tests to parseDN (fuzzing, roundtrip ...)

dn.go

@@ -1,13 +1,14 @@
 package ldap
 
 import (
-	"bytes"
 	"encoding/asn1"
 	"encoding/hex"
 	"errors"
 	"fmt"
 	"sort"
 	"strings"
+	"unicode"
+	"unicode/utf8"
 )
 
 // AttributeTypeAndValue represents an attributeTypeAndValue from https://tools.ietf.org/html/rfc4514
@@ -34,6 +35,9 @@ func (a *AttributeTypeAndValue) setValue(s string) error {
 	// AttributeValue is represented by an number sign ('#' U+0023)
 	// character followed by the hexadecimal encoding of each of the octets
 	// of the BER encoding of the X.500 AttributeValue.
+	//
+	// WARNING: we only support hex-encoded ASN.1 DER values here, not
+	// BER encoding. This is a deviation from the RFC.
 	if len(s) > 0 && s[0] == '#' {
 		decodedString, err := decodeEncodedString(s[1:])
 		if err != nil {
@@ -56,59 +60,7 @@ func (a *AttributeTypeAndValue) setValue(s string) error {
 // String returns a normalized string representation of this attribute type and
 // value pair which is the lowercase join of the Type and Value with a "=".
 func (a *AttributeTypeAndValue) String() string {
-	return strings.ToLower(a.Type) + "=" + a.encodeValue()
-}
-
-func (a *AttributeTypeAndValue) encodeValue() string {
-	// Normalize the value first.
-	// value := strings.ToLower(a.Value)
-	value := a.Value
-
-	encodedBuf := bytes.Buffer{}
-
-	escapeChar := func(c byte) {
-		encodedBuf.WriteByte('\\')
-		encodedBuf.WriteByte(c)
-	}
-
-	escapeHex := func(c byte) {
-		encodedBuf.WriteByte('\\')
-		encodedBuf.WriteString(hex.EncodeToString([]byte{c}))
-	}
-
-	for i := 0; i < len(value); i++ {
-		char := value[i]
-		if i == 0 && char == ' ' || char == '#' {
-			// Special case leading space or number sign.
-			escapeChar(char)
-			continue
-		}
-		if i == len(value)-1 && char == ' ' {
-			// Special case trailing space.
-			escapeChar(char)
-			continue
-		}
-
-		switch char {
-		case '"', '+', ',', ';', '<', '>', '\\':
-			// Each of these special characters must be escaped.
-			escapeChar(char)
-			continue
-		}
-
-		if char < ' ' || char > '~' {
-			// All special character escapes are handled first
-			// above. All bytes less than ASCII SPACE and all bytes
-			// greater than ASCII TILDE must be hex-escaped.
-			escapeHex(char)
-			continue
-		}
-
-		// Any other character does not require escaping.
-		encodedBuf.WriteByte(char)
-	}
-
-	return encodedBuf.String()
+	return encodeString(foldString(a.Type), false) + "=" + encodeString(a.Value, true)
 }
 
 // RelativeDN represents a relativeDistinguishedName from https://tools.ietf.org/html/rfc4514
@@ -142,16 +94,23 @@ func (d *DN) String() string {
 	return strings.Join(rdns, ",")
 }
 
+func stripLeadingAndTrailingSpaces(inVal string) string {
+	noSpaces := strings.Trim(inVal, " ")
+
+	// Re-add the trailing space if it was an escaped space
+	if len(noSpaces) > 0 && noSpaces[len(noSpaces)-1] == '\\' && inVal[len(inVal)-1] == ' ' {
+		noSpaces = noSpaces + " "
+	}
+
+	return noSpaces
+}
+
 // Remove leading and trailing spaces from the attribute type and value
 // and unescape any escaped characters in these fields
 //
 // decodeString is based on https://github.com/inteon/cert-manager/blob/ed280d28cd02b262c5db46054d88e70ab518299c/pkg/util/pki/internal/dn.go#L170
 func decodeString(str string) (string, error) {
-	s := []rune(strings.TrimSpace(str))
-	// Re-add the trailing space if the last character was an escaped space character
-	if len(s) > 0 && s[len(s)-1] == '\\' && str[len(str)-1] == ' ' {
-		s = append(s, ' ')
-	}
+	s := []rune(stripLeadingAndTrailingSpaces(str))
 
 	builder := strings.Builder{}
 	for i := 0; i < len(s); i++ {
@@ -212,6 +171,65 @@ func decodeString(str string) (string, error) {
 	return builder.String(), nil
 }
 
+// Escape a string according to RFC 4514
+func encodeString(value string, isValue bool) string {
+	builder := strings.Builder{}
+
+	escapeChar := func(c byte) {
+		builder.WriteByte('\\')
+		builder.WriteByte(c)
+	}
+
+	escapeHex := func(c byte) {
+		builder.WriteByte('\\')
+		builder.WriteString(hex.EncodeToString([]byte{c}))
+	}
+
+	// Loop through each byte and escape as necessary.
+	// Runes that take up more than one byte are escaped
+	// byte by byte (since both bytes are non-ASCII).
+	for i := 0; i < len(value); i++ {
+		char := value[i]
+		if i == 0 && (char == ' ' || char == '#') {
+			// Special case leading space or number sign.
+			escapeChar(char)
+			continue
+		}
+		if i == len(value)-1 && char == ' ' {
+			// Special case trailing space.
+			escapeChar(char)
+			continue
+		}
+
+		switch char {
+		case '"', '+', ',', ';', '<', '>', '\\':
+			// Each of these special characters must be escaped.
+			escapeChar(char)
+			continue
+		}
+
+		if !isValue && char == '=' {
+			// Equal signs have to be escaped only in the type part of
+			// the attribute type and value pair.
+			escapeChar(char)
+			continue
+		}
+
+		if char < ' ' || char > '~' {
+			// All special character escapes are handled first
+			// above. All bytes less than ASCII SPACE and all bytes
+			// greater than ASCII TILDE must be hex-escaped.
+			escapeHex(char)
+			continue
+		}
+
+		// Any other character does not require escaping.
+		builder.WriteByte(char)
+	}
+
+	return builder.String()
+}
+
 func decodeEncodedString(str string) (string, error) {
 	decoded, err := hex.DecodeString(str)
 	if err != nil {
@@ -253,6 +271,10 @@ func ParseDN(str string) (*DN, error) {
 		}
 	)
 
+	// Loop through each character in the string and
+	// build up the attribute type and value pairs.
+	// We only check for ascii characters here, which
+	// allows us to iterate over the string byte by byte.
 	for i := 0; i < len(str); i++ {
 		char := str[i]
 		switch {
@@ -420,3 +442,34 @@ func (r *RelativeDN) hasAllAttributesFold(attrs []*AttributeTypeAndValue) bool {
 func (a *AttributeTypeAndValue) EqualFold(other *AttributeTypeAndValue) bool {
 	return strings.EqualFold(a.Type, other.Type) && strings.EqualFold(a.Value, other.Value)
 }
+
+// foldString returns a folded string such that foldString(x) == foldString(y)
+// is identical to bytes.EqualFold(x, y).
+// based on https://go.dev/src/encoding/json/fold.go
+func foldString(s string) string {
+	builder := strings.Builder{}
+	for _, char := range s {
+		// Handle single-byte ASCII.
+		if char < utf8.RuneSelf {
+			if 'A' <= char && char <= 'Z' {
+				char += 'a' - 'A'
+			}
+			builder.WriteRune(char)
+			continue
+		}
+
+		builder.WriteRune(foldRune(char))
+	}
+	return builder.String()
+}
+
+// foldRune is returns the smallest rune for all runes in the same fold set.
+func foldRune(r rune) rune {
+	for {
+		r2 := unicode.SimpleFold(r)
+		if r2 <= r {
+			return r
+		}
+		r = r2
+	}
+}