net/mail: properly handle special characters in phrase and obs-phrase

Fixes a couple of misalignments with RFC 5322 which introduce
significant diffs between (mostly) conformant parsers.

This change reverts the changes made in CL50911, which allowed certain
special RFC 5322 characters to appear unquoted in the "phrase" syntax.
It is unclear why this change was made in the first place, and created
a divergence from comformant parsers. In particular this resulted in
treating comments in display names incorrectly.

Additionally properly handle trailing malformed comments in the group
syntax.

Fixes #65083

Change-Id: I00dddc044c6ae3381154e43236632604c390f672
Reviewed-on: https://go-review.googlesource.com/c/go/+/555596
Reviewed-by: Damien Neil <[email protected]>
LUCI-TryBot-Result: Go LUCI <[email protected]>

Author: rolandshoemaker

src/net/mail/message.go

@@ -280,7 +280,7 @@ func (a *Address) String() string {
 	// Add quotes if needed
 	quoteLocal := false
 	for i, r := range local {
-		if isAtext(r, false, false) {
+		if isAtext(r, false) {
 			continue
 		}
 		if r == '.' {
@@ -444,7 +444,7 @@ func (p *addrParser) parseAddress(handleGroup bool) ([]*Address, error) {
 	if !p.consume('<') {
 		atext := true
 		for _, r := range displayName {
-			if !isAtext(r, true, false) {
+			if !isAtext(r, true) {
 				atext = false
 				break
 			}
@@ -479,7 +479,9 @@ func (p *addrParser) consumeGroupList() ([]*Address, error) {
 	// handle empty group.
 	p.skipSpace()
 	if p.consume(';') {
-		p.skipCFWS()
+		if !p.skipCFWS() {
+			return nil, errors.New("mail: misformatted parenthetical comment")
+		}
 		return group, nil
 	}
 
@@ -496,7 +498,9 @@ func (p *addrParser) consumeGroupList() ([]*Address, error) {
 			return nil, errors.New("mail: misformatted parenthetical comment")
 		}
 		if p.consume(';') {
-			p.skipCFWS()
+			if !p.skipCFWS() {
+				return nil, errors.New("mail: misformatted parenthetical comment")
+			}
 			break
 		}
 		if !p.consume(',') {
@@ -566,6 +570,12 @@ func (p *addrParser) consumePhrase() (phrase string, err error) {
 	var words []string
 	var isPrevEncoded bool
 	for {
+		// obs-phrase allows CFWS after one word
+		if len(words) > 0 {
+			if !p.skipCFWS() {
+				return "", errors.New("mail: misformatted parenthetical comment")
+			}
+		}
 		// word = atom / quoted-string
 		var word string
 		p.skipSpace()
@@ -661,7 +671,6 @@ Loop:
 // If dot is true, consumeAtom parses an RFC 5322 dot-atom instead.
 // If permissive is true, consumeAtom will not fail on:
 // - leading/trailing/double dots in the atom (see golang.org/issue/4938)
-// - special characters (RFC 5322 3.2.3) except '<', '>', ':' and '"' (see golang.org/issue/21018)
 func (p *addrParser) consumeAtom(dot bool, permissive bool) (atom string, err error) {
 	i := 0
 
@@ -672,7 +681,7 @@ Loop:
 		case size == 1 && r == utf8.RuneError:
 			return "", fmt.Errorf("mail: invalid utf-8 in address: %q", p.s)
 
-		case size == 0 || !isAtext(r, dot, permissive):
+		case size == 0 || !isAtext(r, dot):
 			break Loop
 
 		default:
@@ -850,18 +859,13 @@ func (e charsetError) Error() string {
 
 // isAtext reports whether r is an RFC 5322 atext character.
 // If dot is true, period is included.
-// If permissive is true, RFC 5322 3.2.3 specials is included,
-// except '<', '>', ':' and '"'.
-func isAtext(r rune, dot, permissive bool) bool {
+func isAtext(r rune, dot bool) bool {
 	switch r {
 	case '.':
 		return dot
 
 	// RFC 5322 3.2.3. specials
-	case '(', ')', '[', ']', ';', '@', '\\', ',':
-		return permissive
-
-	case '<', '>', '"', ':':
+	case '(', ')', '<', '>', '[', ']', ':', ';', '@', '\\', ',', '"': // RFC 5322 3.2.3. specials
 		return false
 	}
 	return isVchar(r)

src/net/mail/message_test.go

@@ -385,8 +385,11 @@ func TestAddressParsingError(t *testing.T) {
 		13: {"group not closed: [email protected]", "expected comma"},
 		14: {"group: [email protected], [email protected];", "group with multiple addresses"},
 		15: {"john.doe", "missing '@' or angle-addr"},
-		16: {"john.doe@", "no angle-addr"},
+		16: {"john.doe@", "missing '@' or angle-addr"},
 		17: {"John [email protected]", "no angle-addr"},
+		18: {" group: [email protected]; (asd", "misformatted parenthetical comment"},
+		19: {" group: ; (asd", "misformatted parenthetical comment"},
+		20: {`(John) Doe <[email protected]>`, "missing word in phrase:"},
 	}
 
 	for i, tc := range mustErrTestCases {
@@ -436,24 +439,19 @@ func TestAddressParsing(t *testing.T) {
 				Address: "[email protected]",
 			}},
 		},
-		{
-			`"John (middle) Doe" <[email protected]>`,
-			[]*Address{{
-				Name:    "John (middle) Doe",
-				Address: "[email protected]",
-			}},
-		},
+		// Comment in display name
 		{
 			`John (middle) Doe <[email protected]>`,
 			[]*Address{{
-				Name:    "John (middle) Doe",
+				Name:    "John Doe",
 				Address: "[email protected]",
 			}},
 		},
+		// Display name is quoted string, so comment is not a comment
 		{
-			`John !@M@! Doe <[email protected]>`,
+			`"John (middle) Doe" <[email protected]>`,
 			[]*Address{{
-				Name:    "John !@M@! Doe",
+				Name:    "John (middle) Doe",
 				Address: "[email protected]",
 			}},
 		},
@@ -788,6 +786,26 @@ func TestAddressParsing(t *testing.T) {
 				},
 			},
 		},
+		// Comment in group display name
+		{
+			`group (comment:): [email protected], [email protected];`,
+			[]*Address{
+				{
+					Address: "[email protected]",
+				},
+				{
+					Address: "[email protected]",
+				},
+			},
+		},
+		{
+			`x(:"):"@a.example;("@b.example;`,
+			[]*Address{
+				{
+					Address: `@a.example;(@b.example`,
+				},
+			},
+		},
 	}
 	for _, test := range tests {
 		if len(test.exp) == 1 {