test: fix pre-existing storage and audio diarization test failures

These tests failed on the published master independently of the install fix;
clearing them so CI can go green.

- delete_job now returns bool (True if deleted, False if not found) on the
  StorageBackend ABC and the file and sqlite backends, matching test
  expectations and giving callers a not-found signal.
- update two stale diarization test assertions from the removed pyannote
  use_auth_token= kwarg to the current token= kwarg (the pipeline already
  uses token=).
- mock PyAnnotePipeline.from_pretrained in the two modular-registration
  tests so they no longer attempt a real gated-model download.

Full suite: 1026 passed, 24 skipped, coverage 57.66%.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: InfantLab

src/videoannotator/storage/base.py

@@ -94,11 +94,14 @@ def list_jobs(self, status_filter: str | None = None) -> list[str]:
         pass
 
     @abstractmethod
-    def delete_job(self, job_id: str) -> None:
+    def delete_job(self, job_id: str) -> bool:
         """Delete all data for a job.
 
         Args:
             job_id: Unique job identifier
+
+        Returns:
+            True if the job existed and was deleted, False if it was not found.
         """
         pass
 

src/videoannotator/storage/file_backend.py

@@ -198,18 +198,24 @@ def get_all_jobs(self, status_filter: str | None = None) -> list[BatchJob]:
 
         return sorted(jobs, key=lambda j: j.created_at)
 
-    def delete_job(self, job_id: str) -> None:
-        """Delete all data for a job."""
+    def delete_job(self, job_id: str) -> bool:
+        """Delete all data for a job.
+
+        Returns:
+            True if the job existed and was deleted, False if it was not found.
+        """
         job_dir = self._get_job_dir(job_id)
 
         if job_dir.exists():
             import shutil
 
             shutil.rmtree(job_dir)
             self.logger.info(f"Deleted job {job_id}")
-        else:
-            self.logger.warning(f"Job {job_id} not found for deletion")
-            self.logger.warning(f"Job directory not found: {job_dir}")
+            return True
+
+        self.logger.warning(f"Job {job_id} not found for deletion")
+        self.logger.warning(f"Job directory not found: {job_dir}")
+        return False
 
     def get_stats(self) -> dict[str, Any]:
         """Get storage statistics."""

src/videoannotator/storage/sqlite_backend.py

@@ -363,8 +363,12 @@ def list_jobs(self, status_filter: str | None = None) -> list[str]:
             self.logger.error(f"[ERROR] Failed to list jobs: {e}")
             return []
 
-    def delete_job(self, job_id: str) -> None:
-        """Delete all data for a job including persistent storage."""
+    def delete_job(self, job_id: str) -> bool:
+        """Delete all data for a job including persistent storage.
+
+        Returns:
+            True if the job existed and was deleted, False if it was not found.
+        """
         import shutil
 
         from ..storage.config import get_job_storage_path
@@ -397,6 +401,8 @@ def delete_job(self, job_id: str) -> None:
                     f"[WARNING] Failed to delete storage for job {job_id}: {storage_error}"
                 )
 
+            return deleted_count > 0
+
         except SQLAlchemyError as e:
             self.logger.error(f"[ERROR] Failed to delete job {job_id}: {e}")
             raise

tests/pipelines/test_audio_individual_components.py

@@ -88,7 +88,7 @@ def test_diarization_pipeline_initialize_success(self, mock_pyannote):
             or os.environ.get("HUGGINGFACE_TOKEN")
             or "FAKE_TOKEN_FOR_TESTING"
         )
-        assert called_kwargs["use_auth_token"] == expected_token
+        assert called_kwargs["token"] == expected_token
 
         pipeline.cleanup()
 

tests/pipelines/test_audio_pipeline.py

@@ -47,8 +47,16 @@ def test_audio_pipeline_default_config(self):
         assert pipeline.config["pipelines"]["speech_recognition"]["enabled"]
         assert pipeline.config["pipelines"]["speaker_diarization"]["enabled"]
 
-    def test_modular_pipeline_initialization(self):
+    @patch(
+        "videoannotator.pipelines.audio_processing.diarization_pipeline.PYANNOTE_AVAILABLE",
+        True,
+    )
+    @patch(
+        "videoannotator.pipelines.audio_processing.diarization_pipeline.PyAnnotePipeline"
+    )
+    def test_modular_pipeline_initialization(self, mock_pyannote):
         """Test that modular pipeline components are properly initialized."""
+        mock_pyannote.from_pretrained.return_value = Mock()
         config = {
             "pipelines": {
                 "speech_recognition": {"enabled": True, "model": "base"},
@@ -121,8 +129,16 @@ def test_speech_recognition_component(self, mock_whisper, temp_audio_file):
         except Exception as e:
             pytest.skip(f"Speech recognition test failed: {e}")
 
-    def test_speaker_diarization_component(self, temp_audio_file):
+    @patch(
+        "videoannotator.pipelines.audio_processing.diarization_pipeline.PYANNOTE_AVAILABLE",
+        True,
+    )
+    @patch(
+        "videoannotator.pipelines.audio_processing.diarization_pipeline.PyAnnotePipeline"
+    )
+    def test_speaker_diarization_component(self, mock_pyannote, temp_audio_file):
         """Test speaker diarization component within modular pipeline."""
+        mock_pyannote.from_pretrained.return_value = Mock()
         config = {
             "pipelines": {
                 "speech_recognition": {

tests/pipelines/test_audio_speech_pipeline.py

@@ -96,7 +96,7 @@ def test_diarization_pipeline_initialize_success(self, mock_pyannote):
             # Check that model was loaded with correct parameters
             mock_pyannote.from_pretrained.assert_called_once_with(
                 "pyannote/speaker-diarization-3.1",
-                use_auth_token="FAKE_TOKEN_FOR_TESTING",
+                token="FAKE_TOKEN_FOR_TESTING",
             )
 
             pipeline.cleanup()