Support quoted printable and filter out some auto-generated Gmail labels

internal/transfer/provider_mbox_gmail_labels.go

@@ -21,6 +21,7 @@ import (
 	"bufio"
 	"bytes"
 	"io"
+	"mime"
 	"os"
 	"strings"
 )
@@ -29,6 +30,17 @@ type stringSet map[string]bool
 
 const xGmailLabelsHeader = "X-Gmail-Labels"
 
+// filteredOutGmailLabels is set of labels which we don't want to show to users
+// as they might be auto-generated by Gmail and unwanted.
+var filteredOutGmailLabels = []string{ //nolint[gochecknoglobals]
+	"Unread",
+	"Opened",
+	"IMAP_Junk",
+	"IMAP_NonJunk",
+	"IMAP_NotJunk",
+	"IMAP_$NotJunk",
+}
+
 func getGmailLabelsFromMboxFile(filePath string) (stringSet, error) {
 	f, err := os.Open(filePath) //nolint[gosec]
 	if err != nil {
@@ -38,7 +50,7 @@ func getGmailLabelsFromMboxFile(filePath string) (stringSet, error) {
 }
 
 func getGmailLabelsFromMboxReader(f io.Reader) (stringSet, error) {
-	allLabels := map[string]bool{}
+	allLabels := stringSet{}
 
 	// Scanner is not used as it does not support long lines and some mbox
 	// files contain very long lines even though that should not be happening.
@@ -78,12 +90,28 @@ func getGmailLabelsFromMessage(body []byte) (stringSet, error) {
 
 func getGmailLabelsFromValue(value string) stringSet {
 	value = strings.TrimPrefix(value, xGmailLabelsHeader+":")
-	labels := map[string]bool{}
+	if decoded, err := new(mime.WordDecoder).DecodeHeader(value); err != nil {
+		log.WithError(err).Error("Failed to decode header")
+	} else {
+		value = decoded
+	}
+
+	labels := stringSet{}
 	for _, label := range strings.Split(value, ",") {
 		label = strings.TrimSpace(label)
 		if label == "" {
 			continue
 		}
+		skip := false
+		for _, filteredOutLabel := range filteredOutGmailLabels {
+			if label == filteredOutLabel {
+				skip = true
+				break
+			}
+		}
+		if skip {
+			continue
+		}
 		labels[label] = true
 	}
 	return labels

internal/transfer/provider_mbox_gmail_labels_test.go

@@ -114,6 +114,8 @@ func TestGetGmailLabelsFromValue(t *testing.T) {
 		{" , ", toSet()},
 		{" ", toSet()},
 		{"", toSet()},
+		{"=?UTF-8?Q?Archived,Category_personal,test_=F0=9F=98=80=F0=9F=99=83?=", toSet("Archived", "Category personal", "test 😀🙃")},
+		{"IMAP_NotJunk,Foo,Opened,bar,Unread", toSet("Foo", "bar")},
 	}
 	for _, tc := range tests {
 		tc := tc