Fix audio error reporting and probe timeout handling

src/utils/files.py

@@ -11,6 +11,7 @@ from typing import Any, Protocol
 from fastapi import UploadFile
 from nanoid import generate as generate_nanoid
 from openai import APIError
+import sentry_sdk
 from sqlalchemy import Integer, select
 from sqlalchemy.ext.asyncio import AsyncSession
 
@@ -165,6 +166,7 @@ class AudioProcessor:
                 content_type=normalized_content_type,
             )
         except APIError as exc:
+            sentry_sdk.capture_exception(exc)
             raise FileProcessingError("Audio transcription failed") from exc
         return ExtractedFileText(
             text=text,
@@ -231,7 +233,8 @@ class AudioProcessor:
                 "Audio uploads require ffmpeg and ffprobe to be installed on the server"
             ) from exc
         except subprocess.TimeoutExpired as exc:
-            raise ValidationException("Audio validation timed out") from exc
+            sentry_sdk.capture_exception(exc)
+            raise FileProcessingError("Audio validation timed out") from exc
         except (subprocess.CalledProcessError, ValueError) as exc:
             raise ValidationException("Uploaded audio is invalid or unreadable") from exc
 

tests/utils/test_audio_processing.py

@@ -67,14 +67,17 @@ def test_probe_audio_duration_timeout_raises_validation_exception():
     processor = AudioProcessor()
 
     with (
+        patch("src.utils.files.sentry_sdk.capture_exception") as mock_capture,
         patch(
             "src.utils.files.subprocess.run",
             side_effect=subprocess.TimeoutExpired(cmd="ffprobe", timeout=10),
         ),
-        pytest.raises(ValidationException, match="Audio validation timed out"),
+        pytest.raises(FileProcessingError, match="Audio validation timed out"),
     ):
         processor.probe_audio_duration_seconds_from_path(Path("/tmp/audio.mp3"))
 
+    mock_capture.assert_called_once()
+
 
 @pytest.mark.asyncio
 async def test_audio_upload_requires_openai_client_before_processing():
@@ -203,6 +206,7 @@ async def test_audio_processor_extract_text_wraps_provider_errors():
     request = httpx.Request("POST", "https://api.openai.com/v1/audio/transcriptions")
 
     with (
+        patch("src.utils.files.sentry_sdk.capture_exception") as mock_capture,
         patch.object(processor, "_probe_audio_duration_seconds", return_value=1.0),
         patch(
             "src.utils.files.transcribe_audio",
@@ -216,6 +220,8 @@ async def test_audio_processor_extract_text_wraps_provider_errors():
             content_type="audio/mpeg",
         )
 
+    mock_capture.assert_called_once()
+
 
 @pytest.mark.asyncio
 async def test_audio_processor_extract_text_probes_in_background_thread():