improve date parsing for notmuch/maildir

If a message date would fail to parse, the worker would never receive
the MessageInfo it asked for, and so it wouldn't display the message.

The problem is the spec for date formats is too lax, so trying to ensure
we can parse every weird date format out there is not a strategy we want
to pursue. On the other hand, preventing the user from reading and
working with a message due to the error format is also not a solution.

The maildir and notmuch workers will now fallback to the internal date, which
is based on the received header if we can't parse the format of the Date header.

The UI will also fallback to the received header whenever the date header can't
be parsed.

This patch is based on the work done by Lyudmil Angelov <lyudmilangelov@gmail.com>
But tries to handle a parsing error a bit more gracefully instead of just returning
the zero date.

2 files changed, 67 insertions, 26 deletions

diff --git a/lib/format/format.go b/lib/format/format.go
index 5f45e588..cc46da8c 100644
--- a/lib/format/format.go
+++ b/lib/format/format.go
@@ -5,6 +5,7 @@ import (
 	"fmt"
 	gomail "net/mail"
 	"strings"
+	"time"
 	"unicode"
 
 	"git.sr.ht/~sircmpwn/aerc/models"
@@ -111,13 +112,21 @@ func ParseMessageFormat(
 			retval = append(retval, 'd')
 			args = append(args, number)
 		case 'd':
+			date := msg.Envelope.Date
+			if date.IsZero() {
+				date = msg.InternalDate
+			}
 			retval = append(retval, 's')
 			args = append(args,
-				msg.InternalDate.Format(timestampformat))
+				dummyIfZeroDate(date, timestampformat))
 		case 'D':
+			date := msg.Envelope.Date
+			if date.IsZero() {
+				date = msg.InternalDate
+			}
 			retval = append(retval, 's')
 			args = append(args,
-				msg.InternalDate.Local().Format(timestampformat))
+				dummyIfZeroDate(date, timestampformat))
 		case 'f':
 			if msg.Envelope == nil {
 				return "", nil,
@@ -335,3 +344,10 @@ handle_end_error:
 	return "", nil,
 		errors.New("reached end of string while parsing message format")
 }
+
+func dummyIfZeroDate(date time.Time, format string) string {
+	if date.IsZero() {
+		return strings.Repeat("?", len(format))
+	}
+	return date.Format(format)
+}
diff --git a/worker/lib/parse.go b/worker/lib/parse.go
index a14a6d33..58327c9b 100644
--- a/worker/lib/parse.go
+++ b/worker/lib/parse.go
@@ -102,11 +102,9 @@ func ParseEntityStructure(e *message.Entity) (*models.BodyStructure, error) {
 	return &body, nil
 }
 
+var DateParseError = errors.New("date parsing failed")
+
 func parseEnvelope(h *mail.Header) (*models.Envelope, error) {
-	date, err := parseDate(h)
-	if err != nil {
-		return nil, fmt.Errorf("could not parse date header: %v", err)
-	}
 	from, err := parseAddressList(h, "from")
 	if err != nil {
 		return nil, fmt.Errorf("could not read from address: %v", err)
@@ -135,6 +133,12 @@ func parseEnvelope(h *mail.Header) (*models.Envelope, error) {
 	if err != nil {
 		return nil, fmt.Errorf("could not read message id: %v", err)
 	}
+	date, err := parseDate(h)
+	if err != nil {
+		// still return a valid struct plus a sentinel date parsing error
+		// if only the date parsing failed
+		err = fmt.Errorf("%w: %v", DateParseError, err)
+	}
 	return &models.Envelope{
 		Date:      date,
 		Subject:   subj,
@@ -144,28 +148,24 @@ func parseEnvelope(h *mail.Header) (*models.Envelope, error) {
 		To:        to,
 		Cc:        cc,
 		Bcc:       bcc,
-	}, nil
+	}, err
 }
 
-// parseDate extends the built-in date parser with additional layouts which are
-// non-conforming but appear in the wild.
+// parseDate tries to parse the date from the Date header with non std formats
+// if this fails it tries to parse the received header as well
 func parseDate(h *mail.Header) (time.Time, error) {
-	t, parseErr := h.Date()
-	if parseErr == nil {
+	t, err := h.Date()
+	if err == nil {
 		return t, nil
 	}
 	text, err := h.Text("date")
-	if err != nil {
-		return time.Time{}, errors.New("no date header")
-	}
-	// sometimes, no error occurs but the date is empty. In this case, guess time from received header field
-	if text == "" {
-		guess, err := h.Text("received")
-		if err != nil {
-			return time.Time{}, errors.New("no received header")
+	// sometimes, no error occurs but the date is empty.
+	// In this case, guess time from received header field
+	if err != nil || text == "" {
+		t, err := parseReceivedHeader(h)
+		if err == nil {
+			return t, nil
 		}
-		t, _ := time.Parse(time.RFC1123Z, dateRe.FindString(guess))
-		return t, nil
 	}
 	layouts := []string{
 		// X-Mailer: EarthLink Zoo Mail 1.0
@@ -176,7 +176,21 @@ func parseDate(h *mail.Header) (time.Time, error) {
 			return t, nil
 		}
 	}
-	return time.Time{}, fmt.Errorf("unrecognized date format: %s", text)
+	// still no success, try the received header as a last resort
+	t, err = parseReceivedHeader(h)
+	if err != nil {
+		return time.Time{}, fmt.Errorf("unrecognized date format: %s", text)
+	}
+	return t, nil
+}
+
+func parseReceivedHeader(h *mail.Header) (time.Time, error) {
+	guess, err := h.Text("received")
+	if err != nil {
+		return time.Time{}, fmt.Errorf("received header not parseable: %v",
+			err)
+	}
+	return time.Parse(time.RFC1123Z, dateRe.FindString(guess))
 }
 
 func parseAddressList(h *mail.Header, key string) ([]*models.Address, error) {
@@ -231,9 +245,20 @@ func MessageInfo(raw RawMessage) (*models.MessageInfo, error) {
 	if err != nil {
 		return nil, fmt.Errorf("could not get structure: %v", err)
 	}
-	env, err := parseEnvelope(&mail.Header{msg.Header})
-	if err != nil {
-		return nil, fmt.Errorf("could not get envelope: %v", err)
+	h := &mail.Header{msg.Header}
+	env, err := parseEnvelope(h)
+	if err != nil && !errors.Is(err, DateParseError) {
+		return nil, fmt.Errorf("could not parse envelope: %v", err)
+		// if only the date parsing failed we still get the rest of the
+		// envelop structure in a valid state.
+		// Date parsing errors are fairly common and it's better to be
+		// slightly off than to not be able to read the mails at all
+		// hence we continue here
+	}
+	recDate, _ := parseReceivedHeader(h)
+	if recDate.IsZero() {
+		// better than nothing, if incorrect
+		recDate = env.Date
 	}
 	flags, err := raw.ModelFlags()
 	if err != nil {
@@ -248,7 +273,7 @@ func MessageInfo(raw RawMessage) (*models.MessageInfo, error) {
 		Envelope:      env,
 		Flags:         flags,
 		Labels:        labels,
-		InternalDate:  env.Date,
+		InternalDate:  recDate,
 		RFC822Headers: &mail.Header{msg.Header},
 		Size:          0,
 		Uid:           raw.UID(),