filter: merge added, which merges multiple captions with same timestamps into one

Author: martinlindhe

cmd/subber/subber.go

@@ -21,7 +21,7 @@ var (
 	keepAds     = kingpin.Flag("keep-ads", "Do not strip advertisement captions.").Bool()
 	skipBackups = kingpin.Flag("skip-backups", "Do not make backup (.srt.org) of original .srt").Bool()
 	language    = kingpin.Flag("language", "Language.").Default("en").String()
-	filterName  = kingpin.Flag("filter", "Filter (none, caps, html, ocr, flip, all).").Default("none").String()
+	filterName  = kingpin.Flag("filter", "Filter (none, caps, html, ocr, merge, flip, all).").Default("none").String()
 	sync        = kingpin.Flag("sync", "Synchronize captions (milliseconds).").Int()
 )
 

filter.go

@@ -17,6 +17,8 @@ func (subtitle *Subtitle) FilterCaptions(filter string) {
 		subtitle.filterHTML()
 	case "ocr":
 		subtitle.filterOCR()
+	case "merge":
+		subtitle.filterMerge()
 	case "flip":
 		subtitle.filterFlip()
 	case "none":

filter_merge.go

@@ -0,0 +1,36 @@
+package subtitles
+
+// filterMerge combines all separate captions with same time spawn into one caption
+func (subtitle *Subtitle) filterMerge() *Subtitle {
+
+	idxToRemove := []int{}
+	for i, cap := range subtitle.Captions {
+		for j, pCap := range subtitle.Captions[0:i] {
+			if pCap.Start == cap.Start && pCap.End == cap.End {
+				subtitle.Captions[j].Text = append(pCap.Text, cap.Text...)
+				idxToRemove = append(idxToRemove, i)
+				break
+			}
+		}
+	}
+
+	newCaptions := []Caption{}
+	for i, cap := range subtitle.Captions {
+		if !contains(idxToRemove, i) {
+			newCaptions = append(newCaptions, cap)
+		}
+	}
+
+	subtitle.Captions = newCaptions
+
+	return subtitle
+}
+
+func contains(s []int, e int) bool {
+	for _, a := range s {
+		if a == e {
+			return true
+		}
+	}
+	return false
+}

filter_merge_test.go

@@ -0,0 +1,53 @@
+package subtitles
+
+import (
+	"testing"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestFilterMerge(t *testing.T) {
+	in := `1
+00:01:28,800 --> 00:01:30,166
+I never thought
+
+2
+00:01:28,800 --> 00:01:30,166
+I would be caught up
+
+3
+00:01:30,266 --> 00:01:32,233
+in a story such as this,
+
+4
+00:01:33,266 --> 00:01:35,867
+because I live on the
+
+5
+00:01:33,266 --> 00:01:35,867
+other side of the world.
+`
+
+	sub, err := NewFromSRT(in)
+	assert.Equal(t, nil, err)
+
+	sub = *sub.filterMerge()
+
+	expected := `1
+00:01:28,800 --> 00:01:30,166
+I never thought
+I would be caught up
+
+3
+00:01:30,266 --> 00:01:32,233
+in a story such as this,
+
+4
+00:01:33,266 --> 00:01:35,867
+because I live on the
+other side of the world.
+
+`
+	assert.Equal(t, expected, sub.AsSRT())
+
+}