net/mail: enhance address parser to include support for domain literals

The existing implementation of the mail package conforms to RFC 5322
for parsing mail messages, but it lacks support for domain literals. This
patch addresses this limitation by adding support for domain literals in
the address parser.

The Addr-Spec Specification, defined in RFC 5322 Section 3.4.1,
outlines the format for email addresses:
https://datatracker.ietf.org/doc/html/rfc5322\#section-3.4.1

Fixes #60206

Change-Id: Ic901418325bd1da69e70800d70b87d658b953738
GitHub-Last-Rev: bdda66f3fe098df3a62d803b1f69e63fef6281e3
GitHub-Pull-Request: golang/go#66075
Reviewed-on: https://go-review.googlesource.com/c/go/+/567777
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: David Chase <drchase@google.com>

2 files changed, 107 insertions, 4 deletions

diff --git a/src/net/mail/message.go b/src/net/mail/message.go
index bb40ccd20a..21b075e78a 100644
--- a/src/net/mail/message.go
+++ b/src/net/mail/message.go
@@ -24,6 +24,7 @@ import (
 	"io"
 	"log"
 	"mime"
+	"net"
 	"net/textproto"
 	"strings"
 	"sync"
@@ -553,10 +554,19 @@ func (p *addrParser) consumeAddrSpec() (spec string, err error) {
 	if p.empty() {
 		return "", errors.New("mail: no domain in addr-spec")
 	}
-	// TODO(dsymonds): Handle domain-literal
-	domain, err = p.consumeAtom(true, false)
-	if err != nil {
-		return "", err
+
+	if p.peek() == '[' {
+		// domain-literal
+		domain, err = p.consumeDomainLiteral()
+		if err != nil {
+			return "", err
+		}
+	} else {
+		// dot-atom
+		domain, err = p.consumeAtom(true, false)
+		if err != nil {
+			return "", err
+		}
 	}
 
 	return localPart + "@" + domain, nil
@@ -707,6 +717,48 @@ Loop:
 	return atom, nil
 }
 
+// consumeDomainLiteral parses an RFC 5322 domain-literal at the start of p.
+func (p *addrParser) consumeDomainLiteral() (string, error) {
+	// Skip the leading [
+	if !p.consume('[') {
+		return "", errors.New(`mail: missing "[" in domain-literal`)
+	}
+
+	// Parse the dtext
+	var dtext string
+	for {
+		if p.empty() {
+			return "", errors.New("mail: unclosed domain-literal")
+		}
+		if p.peek() == ']' {
+			break
+		}
+
+		r, size := utf8.DecodeRuneInString(p.s)
+		if size == 1 && r == utf8.RuneError {
+			return "", fmt.Errorf("mail: invalid utf-8 in domain-literal: %q", p.s)
+		}
+		if !isDtext(r) {
+			return "", fmt.Errorf("mail: bad character in domain-literal: %q", r)
+		}
+
+		dtext += p.s[:size]
+		p.s = p.s[size:]
+	}
+
+	// Skip the trailing ]
+	if !p.consume(']') {
+		return "", errors.New("mail: unclosed domain-literal")
+	}
+
+	// Check if the domain literal is an IP address
+	if net.ParseIP(dtext) == nil {
+		return "", fmt.Errorf("mail: invalid IP address in domain-literal: %q", dtext)
+	}
+
+	return "[" + dtext + "]", nil
+}
+
 func (p *addrParser) consumeDisplayNameComment() (string, error) {
 	if !p.consume('(') {
 		return "", errors.New("mail: comment does not start with (")
@@ -912,3 +964,12 @@ func isMultibyte(r rune) bool {
 func isWSP(r rune) bool {
 	return r == ' ' || r == '\t'
 }
+
+// isDtext reports whether r is an RFC 5322 dtext character.
+func isDtext(r rune) bool {
+	// Printable US-ASCII, excluding "[", "]", or "\".
+	if r == '[' || r == ']' || r == '\\' {
+		return false
+	}
+	return isVchar(r)
+}
diff --git a/src/net/mail/message_test.go b/src/net/mail/message_test.go
index 1f2f62afbf..012d51c3df 100644
--- a/src/net/mail/message_test.go
+++ b/src/net/mail/message_test.go
@@ -390,6 +390,10 @@ func TestAddressParsingError(t *testing.T) {
 		18: {" group: null@example.com; (asd", "misformatted parenthetical comment"},
 		19: {" group: ; (asd", "misformatted parenthetical comment"},
 		20: {`(John) Doe <jdoe@machine.example>`, "missing word in phrase:"},
+		21: {"<jdoe@[" + string([]byte{0xed, 0xa0, 0x80}) + "192.168.0.1]>", "invalid utf-8 in domain-literal"},
+		22: {"<jdoe@[[192.168.0.1]>", "bad character in domain-literal"},
+		23: {"<jdoe@[192.168.0.1>", "unclosed domain-literal"},
+		24: {"<jdoe@[256.0.0.1]>", "invalid IP address in domain-literal"},
 	}
 
 	for i, tc := range mustErrTestCases {
@@ -806,6 +810,20 @@ func TestAddressParsing(t *testing.T) {
 				},
 			},
 		},
+		// Domain-literal
+		{
+			`jdoe@[192.168.0.1]`,
+			[]*Address{{
+				Address: "jdoe@[192.168.0.1]",
+			}},
+		},
+		{
+			`John Doe <jdoe@[192.168.0.1]>`,
+			[]*Address{{
+				Name:    "John Doe",
+				Address: "jdoe@[192.168.0.1]",
+			}},
+		},
 	}
 	for _, test := range tests {
 		if len(test.exp) == 1 {
@@ -956,6 +974,20 @@ func TestAddressParser(t *testing.T) {
 				},
 			},
 		},
+		// Domain-literal
+		{
+			`jdoe@[192.168.0.1]`,
+			[]*Address{{
+				Address: "jdoe@[192.168.0.1]",
+			}},
+		},
+		{
+			`John Doe <jdoe@[192.168.0.1]>`,
+			[]*Address{{
+				Name:    "John Doe",
+				Address: "jdoe@[192.168.0.1]",
+			}},
+		},
 	}
 
 	ap := AddressParser{WordDecoder: &mime.WordDecoder{
@@ -1062,6 +1094,15 @@ func TestAddressString(t *testing.T) {
 			&Address{Name: string([]byte{0xed, 0xa0, 0x80}), Address: "invalid-utf8@example.net"},
 			"=?utf-8?q?=ED=A0=80?= <invalid-utf8@example.net>",
 		},
+		// Domain-literal
+		{
+			&Address{Address: "bob@[192.168.0.1]"},
+			"<bob@[192.168.0.1]>",
+		},
+		{
+			&Address{Name: "Bob", Address: "bob@[192.168.0.1]"},
+			`"Bob" <bob@[192.168.0.1]>`,
+		},
 	}
 	for _, test := range tests {
 		s := test.addr.String()
@@ -1115,6 +1156,7 @@ func TestAddressParsingAndFormatting(t *testing.T) {
 		`<"."@example.com>`,
 		`<".."@example.com>`,
 		`<"0:"@0>`,
+		`<Bob@[192.168.0.1]>`,
 	}
 
 	for _, test := range tests {