Added a new JSON public schema as the default output artifact

internal/cli/merge_test.go

@@ -114,6 +114,81 @@ func TestMergeWritesMergedOutputAndReport(t *testing.T) {
 	}
 }
 
+func TestMergeWritesDefaultOutputSchema(t *testing.T) {
+	dir := t.TempDir()
+	input := writeJSONFile(t, dir, "input.json", `{
+		"segments": [
+			{"start": 1, "end": 1.6, "text": "yeah"}
+		]
+	}`)
+	output := filepath.Join(dir, "merged.json")
+
+	err := executeMergeRaw(
+		"--input-file", input,
+		"--output-file", output,
+	)
+	if err != nil {
+		t.Fatalf("merge failed: %v", err)
+	}
+
+	var transcript schema.DefaultTranscript
+	readJSON(t, output, &transcript)
+	if transcript.Metadata.OutputSchema != config.OutputSchemaDefault {
+		t.Fatalf("output_schema = %q, want default", transcript.Metadata.OutputSchema)
+	}
+	if len(transcript.Segments) != 1 {
+		t.Fatalf("segment count = %d, want 1", len(transcript.Segments))
+	}
+	segment := transcript.Segments[0]
+	if len(segment.Categories) != 1 || segment.Categories[0] != "backchannel" {
+		t.Fatalf("categories = %#v, want [backchannel]", segment.Categories)
+	}
+	outputBytes, err := os.ReadFile(output)
+	if err != nil {
+		t.Fatalf("read output: %v", err)
+	}
+	for _, forbidden := range []string{"overlap_groups", "source", "derived_from", "words"} {
+		if strings.Contains(string(outputBytes), forbidden) {
+			t.Fatalf("default output contains %q:\n%s", forbidden, outputBytes)
+		}
+	}
+}
+
+func TestMergeWritesSeriatimOutputSchema(t *testing.T) {
+	dir := t.TempDir()
+	input := writeJSONFile(t, dir, "input.json", `{
+		"segments": [
+			{"start": 1, "end": 1.6, "text": "yeah"}
+		]
+	}`)
+	output := filepath.Join(dir, "merged.json")
+
+	err := executeMergeRaw(
+		"--input-file", input,
+		"--output-file", output,
+		"--output-schema", "seriatim",
+	)
+	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 len(segment.Categories) != 1 || segment.Categories[0] != "backchannel" {
+		t.Fatalf("categories = %#v, want [backchannel]", segment.Categories)
+	}
+	if segment.Source == "" {
+		t.Fatal("expected full output to include source")
+	}
+	if len(transcript.OverlapGroups) != 0 {
+		t.Fatalf("expected no overlap groups, got %d", len(transcript.OverlapGroups))
+	}
+}
+
 func TestMergeWritesMinimalOutputSchema(t *testing.T) {
 	dir := t.TempDir()
 	input := writeJSONFile(t, dir, "input.json", `{
@@ -1916,11 +1991,29 @@ func TestInvalidTimingFails(t *testing.T) {
 }
 
 func executeMerge(args ...string) error {
+	if !hasOutputSchemaFlag(args) {
+		// Most integration tests were written against the full envelope; keep
+		// that behavior unless the caller explicitly asks for another schema.
+		args = append(args, "--output-schema", config.OutputSchemaSeriatim)
+	}
+	return executeMergeRaw(args...)
+}
+
+func executeMergeRaw(args ...string) error {
 	cmd := NewRootCommand()
 	cmd.SetArgs(append([]string{"merge"}, args...))
 	return cmd.Execute()
 }
 
+func hasOutputSchemaFlag(args []string) bool {
+	for _, arg := range args {
+		if arg == "--output-schema" {
+			return true
+		}
+	}
+	return false
+}
+
 func writeJSONFile(t *testing.T, dir string, name string, content string) string {
 	t.Helper()