net/mail: permit more characters in mail headers

We parse mail messages using net/textproto. For golang#53188, we tightened
up the bytes permitted by net/textproto to match RFC 7230.
However, this package uses RFC 5322 which is more permissive.
Restore the permisiveness we used to have, so that older code
continues to work.

Fixes golang#58862
Fixes golang#60332

Change-Id: I5437f5e18a756f6ca61c13c4d8ba727be73eff9a
Reviewed-on: https://go-review.googlesource.com/c/go/+/504416
Run-TryBot: Ian Lance Taylor <[email protected]>
Run-TryBot: Ian Lance Taylor <[email protected]>
TryBot-Result: Gopher Robot <[email protected]>
Reviewed-by: Damien Neil <[email protected]>
Reviewed-by: Ian Lance Taylor <[email protected]>
Auto-Submit: Ian Lance Taylor <[email protected]>

Author: ianlancetaylor

src/net/mail/message.go

@@ -14,6 +14,7 @@ Notable divergences:
     such as breaking addresses across lines.
   - No unicode normalization is performed.
   - The special characters ()[]:;@\, are allowed to appear unquoted in names.
+  - A leading From line is permitted, as in mbox format (RFC 4155).
 */
 package mail
 
@@ -53,7 +54,7 @@ type Message struct {
 func ReadMessage(r io.Reader) (msg *Message, err error) {
 	tp := textproto.NewReader(bufio.NewReader(r))
 
-	hdr, err := tp.ReadMIMEHeader()
+	hdr, err := readHeader(tp)
 	if err != nil && (err != io.EOF || len(hdr) == 0) {
 		return nil, err
 	}
@@ -64,6 +65,54 @@ func ReadMessage(r io.Reader) (msg *Message, err error) {
 	}, nil
 }
 
+// readHeader reads the message headers from r.
+// This is like textproto.ReadMIMEHeader, but doesn't validate.
+// The fix for issue #53188 tightened up net/textproto to enforce
+// restrictions of RFC 7230.
+// This package implements RFC 5322, which does not have those restrictions.
+// This function copies the relevant code from net/textproto,
+// simplified for RFC 5322.
+func readHeader(r *textproto.Reader) (map[string][]string, error) {
+	m := make(map[string][]string)
+
+	// The first line cannot start with a leading space.
+	if buf, err := r.R.Peek(1); err == nil && (buf[0] == ' ' || buf[0] == '\t') {
+		line, err := r.ReadLine()
+		if err != nil {
+			return m, err
+		}
+		return m, errors.New("malformed initial line: " + line)
+	}
+
+	for {
+		kv, err := r.ReadContinuedLine()
+		if kv == "" {
+			return m, err
+		}
+
+		// Key ends at first colon.
+		k, v, ok := strings.Cut(kv, ":")
+		if !ok {
+			return m, errors.New("malformed header line: " + kv)
+		}
+		key := textproto.CanonicalMIMEHeaderKey(k)
+
+		// Permit empty key, because that is what we did in the past.
+		if key == "" {
+			continue
+		}
+
+		// Skip initial spaces in value.
+		value := strings.TrimLeft(v, " \t")
+
+		m[key] = append(m[key], value)
+
+		if err != nil {
+			return m, err
+		}
+	}
+}
+
 // Layouts suitable for passing to time.Parse.
 // These are tried in order.
 var (

src/net/mail/message_test.go

@@ -52,6 +52,34 @@ Version: 1
 		},
 		body: "",
 	},
+	{
+		// RFC 5322 permits any printable ASCII character,
+		// except colon, in a header key. Issue #58862.
+		in: `From: [email protected]
+Custom/Header: v
+
+Body
+`,
+		header: Header{
+			"From":          []string{"[email protected]"},
+			"Custom/Header": []string{"v"},
+		},
+		body: "Body\n",
+	},
+	{
+		// RFC 4155 mbox format. We've historically permitted this,
+		// so we continue to permit it. Issue #60332.
+		in: `From [email protected] Mon Jun 19 00:00:00 2023
+From: [email protected]
+
+Hello, gophers!
+`,
+		header: Header{
+			"From":                               []string{"[email protected]"},
+			"From [email protected] Mon Jun 19 00": []string{"00:00 2023"},
+		},
+		body: "Hello, gophers!\n",
+	},
 }
 
 func TestParsing(t *testing.T) {