Implemented a new internal/danglers package with deterministic two-pass dangling-end then dangling-start resolution

This commit is contained in:
2026-04-28 15:38:16 -05:00
parent 47b6727973
commit f1ce35dfc3
8 changed files with 602 additions and 8 deletions

View File

@@ -95,6 +95,7 @@ func TestMergeWritesMergedOutputAndReport(t *testing.T) {
"resolve-overlaps",
"backchannel",
"filler",
"resolve-danglers",
"coalesce",
"detect-overlaps",
"autocorrect",
@@ -701,6 +702,128 @@ func TestMergeCoalesceGapOverridePreventsMerge(t *testing.T) {
}
}
func TestMergeResolveDanglersMergesDanglingEnd(t *testing.T) {
dir := t.TempDir()
inputA := writeJSONFile(t, dir, "a.json", `{
"segments": [
{
"start": 1,
"end": 4,
"text": "main tail.",
"words": [
{"word": "main", "start": 1, "end": 1.1},
{"word": "tail.", "start": 3, "end": 3.1}
]
}
]
}`)
inputB := writeJSONFile(t, dir, "b.json", `{
"segments": [
{
"start": 1.5,
"end": 2,
"text": "interruption",
"words": [
{"word": "interruption", "start": 1.5, "end": 2}
]
}
]
}`)
speakers := writeYAMLFile(t, dir, "speakers.yml", `match:
- speaker: Alice
match: ["a.json"]
- speaker: Bob
match: ["b.json"]
`)
output := filepath.Join(dir, "merged.json")
reportPath := filepath.Join(dir, "report.json")
err := executeMerge(
"--input-file", inputA,
"--input-file", inputB,
"--speakers", speakers,
"--output-file", output,
"--report-file", reportPath,
)
if err != nil {
t.Fatalf("merge failed: %v", err)
}
var transcript model.FinalTranscript
readJSON(t, output, &transcript)
if transcript.Segments[0].Speaker != "Alice" || transcript.Segments[0].Text != "main tail." {
t.Fatalf("first segment = %#v, want Alice merged dangling end", transcript.Segments[0])
}
if transcript.Segments[0].ID != 1 || transcript.Segments[1].ID != 2 {
t.Fatalf("ids not sequential after resolve-danglers: %#v", transcript.Segments)
}
var rpt report.Report
readJSON(t, reportPath, &rpt)
if !hasReportEvent(rpt, "postprocessing", "resolve-danglers", "merged 1 dangling segment(s) into 1 target segment(s)") {
t.Fatal("expected resolve-danglers report event")
}
}
func TestMergeResolveDanglersMergesDanglingStart(t *testing.T) {
dir := t.TempDir()
inputA := writeJSONFile(t, dir, "a.json", `{
"segments": [
{
"start": 1,
"end": 4,
"text": "start target words",
"words": [
{"word": "start", "start": 1, "end": 1.1},
{"word": "target", "start": 3, "end": 3.1},
{"word": "words", "start": 3.2, "end": 3.3}
]
}
]
}`)
inputB := writeJSONFile(t, dir, "b.json", `{
"segments": [
{
"start": 1.5,
"end": 2,
"text": "interruption",
"words": [
{"word": "interruption", "start": 1.5, "end": 2}
]
}
]
}`)
speakers := writeYAMLFile(t, dir, "speakers.yml", `match:
- speaker: Alice
match: ["a.json"]
- speaker: Bob
match: ["b.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)
found := false
for _, segment := range transcript.Segments {
if segment.Speaker == "Alice" && segment.Text == "start target words" {
found = true
}
}
if !found {
t.Fatalf("expected resolved dangling start in output, got %#v", transcript.Segments)
}
}
func TestMergeTagsBackchannelSegments(t *testing.T) {
dir := t.TempDir()
input := writeJSONFile(t, dir, "input.json", `{