feat(imapx): message dedup key (Message-ID + header fallback)

go.mod

@@ -2,7 +2,10 @@ module github.com/vasyansk/imap-copier
 
 go 1.22.0
 
-require github.com/jackc/pgx/v5 v5.7.0
+require (
+	github.com/emersion/go-imap/v2 v2.0.0-beta.8
+	github.com/jackc/pgx/v5 v5.7.0
+)
 
 require (
 	github.com/jackc/pgpassfile v1.0.0 // indirect

go.sum

@@ -1,6 +1,8 @@
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/emersion/go-imap/v2 v2.0.0-beta.8 h1:5IXZK1E33DyeP526320J3RS7eFlCYGFgtbrfapqDPug=
+github.com/emersion/go-imap/v2 v2.0.0-beta.8/go.mod h1:dhoFe2Q0PwLrMD7oZw8ODuaD0vLYPe5uj2wcOMnvh48=
 github.com/jackc/pgpassfile v1.0.0 h1:/6Hmqy13Ss2zCq62VdNG8tM1wchn8zjSGOBJ6icpsIM=
 github.com/jackc/pgpassfile v1.0.0/go.mod h1:CEx0iS5ambNFdcRtxPj5JhEz+xB6uRky5eyVu/W2HEg=
 github.com/jackc/pgservicefile v0.0.0-20240606120523-5a60cdf6a761 h1:iCEnooe7UlwOQYpKFhBabPMi4aNAfoODPEFNiAnClxo=

internal/imapx/messagekey.go

@@ -0,0 +1,37 @@
+package imapx
+
+import (
+	"crypto/md5"
+	"fmt"
+	"strings"
+
+	"github.com/emersion/go-imap/v2"
+)
+
+func MessageKey(env *imap.Envelope, size int64) string {
+	if env != nil && env.MessageID != "" {
+		return env.MessageID
+	}
+	var b strings.Builder
+	if env != nil {
+		b.WriteString(addrList(env.From))
+		b.WriteByte('|')
+		b.WriteString(addrList(env.To))
+		b.WriteByte('|')
+		b.WriteString(env.Subject)
+		b.WriteByte('|')
+		b.WriteString(env.Date.UTC().Format("2006-01-02T15:04:05Z"))
+	}
+	b.WriteByte('|')
+	fmt.Fprintf(&b, "%d", size)
+	sum := md5.Sum([]byte(b.String()))
+	return fmt.Sprintf("h:%x", sum)
+}
+
+func addrList(addrs []imap.Address) string {
+	parts := make([]string, 0, len(addrs))
+	for _, a := range addrs {
+		parts = append(parts, a.Mailbox+"@"+a.Host)
+	}
+	return strings.Join(parts, ",")
+}

internal/imapx/messagekey_test.go

@@ -0,0 +1,32 @@
+package imapx
+
+import (
+	"testing"
+	"time"
+
+	"github.com/emersion/go-imap/v2"
+)
+
+func TestMessageKeyPrefersMessageID(t *testing.T) {
+	env := &imap.Envelope{MessageID: "<abc@host>"}
+	if got := MessageKey(env, 100); got != "<abc@host>" {
+		t.Fatalf("got %q, want <abc@host>", got)
+	}
+}
+
+func TestMessageKeyFallbackStable(t *testing.T) {
+	env := &imap.Envelope{
+		Subject: "Hi",
+		Date:    time.Unix(1700000000, 0).UTC(),
+		From:    []imap.Address{{Mailbox: "a", Host: "x.com"}},
+		To:      []imap.Address{{Mailbox: "b", Host: "y.com"}},
+	}
+	k1 := MessageKey(env, 42)
+	k2 := MessageKey(env, 42)
+	if k1 != k2 {
+		t.Fatal("fallback key must be deterministic")
+	}
+	if MessageKey(env, 43) == k1 {
+		t.Fatal("different size must change key")
+	}
+}