Bugfix in the coalesce module

2026-04-27 20:29:55 -05:00
parent fb0519c561
commit 8a95dba276
4 changed files with 248 additions and 6 deletions
--- a/internal/cli/merge_test.go
+++ b/internal/cli/merge_test.go
@@ -665,6 +665,94 @@ func TestMergeTagsFillerSegments(t *testing.T) {
 	}
 }

+func TestMergeCoalescesSameSpeakerBackchannelWithFollowingSegment(t *testing.T) {
+	dir := t.TempDir()
+	input := writeJSONFile(t, dir, "zach.json", `{
+		"segments": [
+			{"start": 1, "end": 1.7, "text": "That makes sense."},
+			{"start": 1.72, "end": 4, "text": "So, like, next thought."}
+		]
+	}`)
+	output := filepath.Join(dir, "merged.json")
+
+	err := executeMerge(
+		"--input-file", input,
+		"--output-file", output,
+	)
+	if err != nil {
+		t.Fatalf("merge failed: %v", err)
+	}
+
+	var transcript model.FinalTranscript
+	readJSON(t, output, &transcript)
+	if len(transcript.Segments) != 1 {
+		t.Fatalf("segment count = %d, want 1", len(transcript.Segments))
+	}
+	segment := transcript.Segments[0]
+	if segment.Text != "That makes sense. So, like, next thought." {
+		t.Fatalf("text = %q", segment.Text)
+	}
+	if segment.SourceRef != "coalesce:1" {
+		t.Fatalf("source_ref = %q, want coalesce:1", segment.SourceRef)
+	}
+	if !equalStrings(segment.DerivedFrom, []string{input + "#0", input + "#1"}) {
+		t.Fatalf("derived_from = %v", segment.DerivedFrom)
+	}
+	if len(segment.Categories) != 0 {
+		t.Fatalf("categories = %v, want none", segment.Categories)
+	}
+}
+
+func TestMergeCoalescesBackchannelAfterDifferentSpeakerIntoFollowingSameSpeakerSegment(t *testing.T) {
+	dir := t.TempDir()
+	inputA := writeJSONFile(t, dir, "mike.json", `{
+		"segments": [
+			{"start": 1, "end": 2, "text": "previous speaker"}
+		]
+	}`)
+	inputB := writeJSONFile(t, dir, "zach.json", `{
+		"segments": [
+			{"start": 2.5, "end": 3, "text": "That makes sense."},
+			{"start": 3.02, "end": 6, "text": "So, like, next thought."}
+		]
+	}`)
+	speakers := writeYAMLFile(t, dir, "speakers.yml", `match:
+  - speaker: Mike
+    match: ["mike.json"]
+  - speaker: Zach
+    match: ["zach.json"]
+`)
+	output := filepath.Join(dir, "merged.json")
+
+	err := executeMerge(
+		"--input-file", inputA,
+		"--input-file", inputB,
+		"--speakers", speakers,
+		"--output-file", output,
+	)
+	if err != nil {
+		t.Fatalf("merge failed: %v", err)
+	}
+
+	var transcript model.FinalTranscript
+	readJSON(t, output, &transcript)
+	if len(transcript.Segments) != 2 {
+		t.Fatalf("segment count = %d, want 2", len(transcript.Segments))
+	}
+	if transcript.Segments[0].Speaker != "Mike" || transcript.Segments[0].Text != "previous speaker" {
+		t.Fatalf("first segment = %#v, want Mike original", transcript.Segments[0])
+	}
+	if transcript.Segments[1].Speaker != "Zach" || transcript.Segments[1].Text != "That makes sense. So, like, next thought." {
+		t.Fatalf("second segment = %#v, want coalesced Zach", transcript.Segments[1])
+	}
+	if !equalStrings(transcript.Segments[1].DerivedFrom, []string{inputB + "#0", inputB + "#1"}) {
+		t.Fatalf("derived_from = %v", transcript.Segments[1].DerivedFrom)
+	}
+	if len(transcript.Segments[1].Categories) != 0 {
+		t.Fatalf("categories = %v, want none", transcript.Segments[1].Categories)
+	}
+}
+
 func TestMergeCoalescesAroundDifferentSpeakerBackchannel(t *testing.T) {
 	dir := t.TempDir()
 	inputA := writeJSONFile(t, dir, "a.json", `{