Verify WhisperX transport replay

internal/app/whisperx_wiring_test.go

@@ -7,9 +7,12 @@ import (
 	"net/http/httptest"
 	"os"
 	"path/filepath"
+	"slices"
 	"strings"
+	"sync"
 	"sync/atomic"
 	"testing"
+	"time"
 
 	"gitea.maximumdirect.net/eric/narratio/internal/adapters/whisperx"
 	"gitea.maximumdirect.net/eric/narratio/internal/config"
@@ -118,6 +121,86 @@ func TestExecuteStagesDefaultWiringWhisperXFailureMarksManifestFailed(t *testing
 	}
 }
 
+func TestExhaustedWhisperXFailurePreservesEarlierSpeakerCheckpoint(t *testing.T) {
+	var successful atomic.Bool
+	var requestMu sync.Mutex
+	var requested []string
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		file, header, err := r.FormFile("file")
+		if err != nil {
+			t.Errorf("FormFile(file) error = %v", err)
+			http.Error(w, "invalid multipart", http.StatusBadRequest)
+			return
+		}
+		_, readErr := io.Copy(io.Discard, file)
+		_ = file.Close()
+		if readErr != nil {
+			t.Errorf("read uploaded file: %v", readErr)
+			return
+		}
+		speaker := strings.TrimSuffix(header.Filename, filepath.Ext(header.Filename))
+		requestMu.Lock()
+		requested = append(requested, speaker)
+		requestMu.Unlock()
+		if speaker == "b" && !successful.Load() {
+			http.Error(w, "temporary", http.StatusInternalServerError)
+			return
+		}
+		_, _ = w.Write([]byte(`{"segments":[]}`))
+	}))
+	defer srv.Close()
+
+	cfg, store := setupTranscribeRecoveryFixture(t, []string{"a", "b", "c"})
+	setWhisperXConfig(cfg, srv.URL, 1)
+	concurrency := 1
+	cfg.Pipeline.WhisperX.Concurrency = &concurrency
+	httpClient, err := whisperx.NewHTTPClient(whisperx.HTTPClientConfig{
+		TranscribeURL: srv.URL,
+		Language:      "en",
+		Timeout:       2 * time.Second,
+		Retries:       1,
+		RetryDelay:    time.Millisecond,
+	})
+	if err != nil {
+		t.Fatalf("NewHTTPClient() error = %v", err)
+	}
+	transcribe := mustSelectStage(t, "transcribe")
+
+	if _, err := executeStages(context.Background(), cfg, []stage.Stage{transcribe}, RunOptions{
+		Env: &Env{ManifestStore: store, WhisperX: httpClient},
+	}); err == nil {
+		t.Fatal("executeStages() error = nil, want exhausted HTTP failure")
+	}
+	failed := loadTranscribeRecoveryManifest(t, store, cfg)
+	if got := transcribeSpeakerKeys(failed); !slices.Equal(got, []string{"a"}) {
+		t.Fatalf("checkpointed speakers = %#v, want a", got)
+	}
+	requestMu.Lock()
+	firstRequests := append([]string(nil), requested...)
+	requested = nil
+	requestMu.Unlock()
+	if !slices.Equal(firstRequests, []string{"a", "b", "b"}) {
+		t.Fatalf("initial HTTP speakers = %#v, want a then two attempts for b", firstRequests)
+	}
+
+	successful.Store(true)
+	if _, err := executeStages(context.Background(), cfg, []stage.Stage{transcribe}, RunOptions{
+		Env: &Env{ManifestStore: store, WhisperX: httpClient},
+	}); err != nil {
+		t.Fatalf("retry executeStages() error = %v", err)
+	}
+	requestMu.Lock()
+	retryRequests := append([]string(nil), requested...)
+	requestMu.Unlock()
+	if !slices.Equal(retryRequests, []string{"b", "c"}) {
+		t.Fatalf("retry HTTP speakers = %#v, want only b and c", retryRequests)
+	}
+	completed := loadTranscribeRecoveryManifest(t, store, cfg)
+	if got := transcribeSpeakerKeys(completed); !slices.Equal(got, []string{"a", "b", "c"}) {
+		t.Fatalf("completed speakers = %#v, want all speakers", got)
+	}
+}
+
 func TestExecuteStagesExplicitWhisperXInjectionOverridesDefaultWiring(t *testing.T) {
 	cfg := testConfig(t)
 	setWhisperXConfig(cfg, "https://127.0.0.1:1/transcribe", 0)