feat: Enhance audio normalization service to handle invalid analysis values and improve fallback logic

app/services/sound_normalizer.py

@@ -216,6 +216,18 @@ class SoundNormalizerService:
 
             logger.debug("Found JSON match: %s", json_match.group())
             analysis_data = json.loads(json_match.group())
+            
+            # Check for invalid values that would cause second pass to fail
+            invalid_values = ["-inf", "inf", "nan"]
+            for key in ["input_i", "input_lra", "input_tp", "input_thresh", "target_offset"]:
+                if str(analysis_data.get(key, "")).lower() in invalid_values:
+                    logger.warning(
+                        "Invalid analysis value for %s: %s. Falling back to one-pass normalization.", 
+                        key, analysis_data.get(key)
+                    )
+                    # Fall back to one-pass normalization
+                    await self._normalize_audio_one_pass(input_path, output_path)
+                    return
 
             # Second pass: normalize with measured values
             logger.debug("Second pass: normalizing %s with measured values", input_path)

tests/services/test_sound_scanner.py

@@ -41,7 +41,7 @@ class TestSoundScannerService:
 
         try:
             hash_value = scanner_service.get_file_hash(temp_path)
-            assert len(hash_value) == 32  # MD5 hash length
+            assert len(hash_value) == 64  # SHA-256 hash length
             assert isinstance(hash_value, str)
         finally:
             temp_path.unlink()
@@ -77,9 +77,7 @@ class TestSoundScannerService:
     @patch("app.services.sound_scanner.ffmpeg.probe")
     def test_get_audio_duration_success(self, mock_probe, scanner_service):
         """Test successful audio duration extraction."""
-        mock_probe.return_value = {
+        mock_probe.return_value = {"format": {"duration": "123.456"}}
-            "format": {"duration": "123.456"}
-        }
 
         temp_path = Path("/fake/path/test.mp3")
         duration = scanner_service.get_audio_duration(temp_path)
@@ -137,8 +135,13 @@ class TestSoundScannerService:
 
         try:
             results = {
-                "scanned": 0, "added": 0, "updated": 0, "deleted": 0,
+                "scanned": 0,
-                "skipped": 0, "errors": 0, "files": []
+                "added": 0,
+                "updated": 0,
+                "deleted": 0,
+                "skipped": 0,
+                "errors": 0,
+                "files": [],
             }
             await scanner_service._sync_audio_file(
                 temp_path, "SDB", existing_sound, results
@@ -153,7 +156,7 @@ class TestSoundScannerService:
         finally:
             temp_path.unlink()
 
-    @pytest.mark.asyncio 
+    @pytest.mark.asyncio
     async def test_sync_audio_file_new(self, scanner_service):
         """Test syncing a new audio file."""
         created_sound = Sound(
@@ -178,8 +181,13 @@ class TestSoundScannerService:
 
         try:
             results = {
-                "scanned": 0, "added": 0, "updated": 0, "deleted": 0,
+                "scanned": 0,
-                "skipped": 0, "errors": 0, "files": []
+                "added": 0,
+                "updated": 0,
+                "deleted": 0,
+                "skipped": 0,
+                "errors": 0,
+                "files": [],
             }
             await scanner_service._sync_audio_file(temp_path, "SDB", None, results)
 
@@ -188,7 +196,7 @@ class TestSoundScannerService:
             assert results["updated"] == 0
             assert len(results["files"]) == 1
             assert results["files"][0]["status"] == "added"
-            
+
             # Verify sound_repo.create was called with correct data
             call_args = scanner_service.sound_repo.create.call_args[0][0]
             assert call_args["type"] == "SDB"
@@ -200,7 +208,7 @@ class TestSoundScannerService:
         finally:
             temp_path.unlink()
 
-    @pytest.mark.asyncio 
+    @pytest.mark.asyncio
     async def test_sync_audio_file_updated(self, scanner_service):
         """Test syncing a file that was modified (different hash)."""
         # Existing sound with different hash than file
@@ -210,7 +218,7 @@ class TestSoundScannerService:
             name="Old Sound",
             filename="test.mp3",
             duration=60000,  # Old duration
-            size=512,      # Old size
+            size=512,  # Old size
             hash="old_hash",  # Old hash
         )
 
@@ -227,8 +235,13 @@ class TestSoundScannerService:
 
         try:
             results = {
-                "scanned": 0, "added": 0, "updated": 0, "deleted": 0,
+                "scanned": 0,
-                "skipped": 0, "errors": 0, "files": []
+                "added": 0,
+                "updated": 0,
+                "deleted": 0,
+                "skipped": 0,
+                "errors": 0,
+                "files": [],
             }
             await scanner_service._sync_audio_file(
                 temp_path, "SDB", existing_sound, results
@@ -240,7 +253,7 @@ class TestSoundScannerService:
             assert len(results["files"]) == 1
             assert results["files"][0]["status"] == "updated"
             assert results["files"][0]["reason"] == "file was modified"
-            
+
             # Verify sound_repo.update was called with correct data
             call_args = scanner_service.sound_repo.update.call_args[0][1]  # update_data
             assert call_args["duration"] == 120000
@@ -251,7 +264,7 @@ class TestSoundScannerService:
         finally:
             temp_path.unlink()
 
-    @pytest.mark.asyncio 
+    @pytest.mark.asyncio
     async def test_sync_audio_file_custom_type(self, scanner_service):
         """Test syncing file with custom type."""
         created_sound = Sound(
@@ -276,8 +289,13 @@ class TestSoundScannerService:
 
         try:
             results = {
-                "scanned": 0, "added": 0, "updated": 0, "deleted": 0,
+                "scanned": 0,
-                "skipped": 0, "errors": 0, "files": []
+                "added": 0,
+                "updated": 0,
+                "deleted": 0,
+                "skipped": 0,
+                "errors": 0,
+                "files": [],
             }
             await scanner_service._sync_audio_file(temp_path, "CUSTOM", None, results)
 
@@ -285,7 +303,7 @@ class TestSoundScannerService:
             assert results["skipped"] == 0
             assert len(results["files"]) == 1
             assert results["files"][0]["status"] == "added"
-            
+
             # Verify sound_repo.create was called with correct data for custom type
             call_args = scanner_service.sound_repo.create.call_args[0][0]
             assert call_args["type"] == "CUSTOM"
@@ -293,6 +311,8 @@ class TestSoundScannerService:
             assert call_args["duration"] == 60000  # Duration in ms
             assert call_args["size"] == 2048
             assert call_args["hash"] == "custom_hash"
-            assert call_args["is_deletable"] is False  # All sounds are set to not deletable
+            assert (
+                call_args["is_deletable"] is False
+            )  # All sounds are set to not deletable
         finally:
-            temp_path.unlink()
+            temp_path.unlink()