dedupe: Add dedupe largest functionality - fixes #2269

cmd/dedupe/dedupe.go

@@ -90,6 +90,7 @@ Dedupe can be run non interactively using the ` + "`" + `--dedupe-mode` + "`" +
   * ` + "`" + `--dedupe-mode first` + "`" + ` - removes identical files then keeps the first one.
   * ` + "`" + `--dedupe-mode newest` + "`" + ` - removes identical files then keeps the newest one.
   * ` + "`" + `--dedupe-mode oldest` + "`" + ` - removes identical files then keeps the oldest one.
+  * ` + "`" + `--dedupe-mode largest` + "`" + ` - removes identical files then keeps the largest one.
   * ` + "`" + `--dedupe-mode rename` + "`" + ` - removes identical files then renames the rest to be different.
 
 For example to rename all the identically named photos in your Google Photos directory, do

fs/operations/dedupe.go

@@ -119,6 +119,7 @@ const (
 	DeduplicateNewest                             // choose the newest object
 	DeduplicateOldest                             // choose the oldest object
 	DeduplicateRename                             // rename the objects
+	DeduplicateLargest                            // choose the largest object
 )
 
 func (x DeduplicateMode) String() string {
@@ -135,6 +136,8 @@ func (x DeduplicateMode) String() string {
 		return "oldest"
 	case DeduplicateRename:
 		return "rename"
+	case DeduplicateLargest:
+		return "largest"
 	}
 	return "unknown"
 }
@@ -154,6 +157,8 @@ func (x *DeduplicateMode) Set(s string) error {
 		*x = DeduplicateOldest
 	case "rename":
 		*x = DeduplicateRename
+	case "largest":
+		*x = DeduplicateLargest
 	default:
 		return errors.Errorf("Unknown mode for dedupe %q.", s)
 	}
@@ -260,6 +265,7 @@ func Deduplicate(f fs.Fs, mode DeduplicateMode) error {
 	if err != nil {
 		return err
 	}
+
 	for remote, objs := range files {
 		if len(objs) > 1 {
 			fs.Logf(remote, "Found %d duplicates - deleting identical copies", len(objs))
@@ -281,6 +287,17 @@ func Deduplicate(f fs.Fs, mode DeduplicateMode) error {
 				dedupeDeleteAllButOne(0, remote, objs)
 			case DeduplicateRename:
 				dedupeRename(remote, objs)
+			case DeduplicateLargest:
+				size, largest, largestIndex := int64(0), int64(-1), -1
+				for i, obj := range objs {
+					size = obj.Size()
+					if size > largest {
+						largest, largestIndex = size, i
+					}
+				}
+				if largestIndex > -1 {
+					dedupeDeleteAllButOne(largestIndex, remote, objs)
+				}
 			case DeduplicateSkip:
 				// skip
 			default:

fs/operations/dedupe_test.go

@@ -131,6 +131,22 @@ func TestDeduplicateOldest(t *testing.T) {
 	fstest.CheckItems(t, r.Fremote, file1)
 }
 
+func TestDeduplicateLargest(t *testing.T) {
+	r := fstest.NewRun(t)
+	defer r.Finalise()
+	skipIfCantDedupe(t, r.Fremote)
+
+	file1 := r.WriteUncheckedObject("one", "This is one", t1)
+	file2 := r.WriteUncheckedObject("one", "This is one too", t2)
+	file3 := r.WriteUncheckedObject("one", "This is another one", t3)
+	r.CheckWithDuplicates(t, file1, file2, file3)
+
+	err := operations.Deduplicate(r.Fremote, operations.DeduplicateLargest)
+	require.NoError(t, err)
+
+	fstest.CheckItems(t, r.Fremote, file3)
+}
+
 func TestDeduplicateRename(t *testing.T) {
 	r := fstest.NewRun(t)
 	defer r.Finalise()