Add cleanup of temp and upload files after transcription job

scraibe/tasks.py

@@ -17,6 +17,19 @@ from .email_sender import create_transcript_docx, create_summary_docx
 logger = logging.getLogger("scraibe.tasks")
 
 
+def _remove_file(path: str):
+    """
+    Remove a file if it exists. Best-effort; logs but never raises.
+    """
+    if not path:
+        return
+    try:
+        if os.path.exists(path):
+            os.remove(path)
+    except Exception as e:
+        logger.warning("Failed to remove file %s: %s", path, e)
+
+
 def get_queue_position(task_id: str) -> int:
     """
     Estimate the job's position in the queue.
@@ -200,6 +213,7 @@ def process_transcription_task(
 ):
     """
     Async task: transcribe audio, optionally summarize, then email results.
+    Cleans up temporary files after completion.
     """
     task_id = self.request.id
 
@@ -207,6 +221,10 @@ def process_transcription_task(
     log_level = os.getenv("LOG_LEVEL", "INFO")
     setup_logging(level=log_level)
 
+    # Track all temporary files to clean up later
+    temp_files = []
+
+    try:
         # 1) Determine queue position and send initial email
         queue_pos = get_queue_position(task_id)
         send_initial_email(to=email_to, queue_pos=queue_pos)
@@ -222,7 +240,6 @@ def process_transcription_task(
             )
             raise
 
-    try:
         # 3) Perform transcription
         if task_type == "transcript_and_summarize":
             result = scraibe.transcript_and_summarize(
@@ -258,11 +275,13 @@ def process_transcription_task(
             f.write("# Transcript\n\n")
             f.write(transcript_text)
         attachments.append(md_transcript_path)
+        temp_files.append(md_transcript_path)
 
         # Transcript as .docx
         docx_transcript_path = tempfile.mktemp(suffix=".docx")
         create_transcript_docx(transcript_text, docx_transcript_path)
         attachments.append(docx_transcript_path)
+        temp_files.append(docx_transcript_path)
 
         # JSON with diarization
         json_data = {
@@ -285,6 +304,7 @@ def process_transcription_task(
         with open(json_path, "w", encoding="utf-8") as f:
             json.dump(json_data, f, indent=2, ensure_ascii=False)
         attachments.append(json_path)
+        temp_files.append(json_path)
 
         # Summary as .md (only when summary is available)
         if summary_text:
@@ -293,11 +313,13 @@ def process_transcription_task(
                 f.write("# Summary\n\n")
                 f.write(summary_text)
             attachments.append(md_summary_path)
+            temp_files.append(md_summary_path)
 
             # Summary as .docx
             docx_summary_path = tempfile.mktemp(suffix=".docx")
             create_summary_docx(summary_text, docx_summary_path)
             attachments.append(docx_summary_path)
+            temp_files.append(docx_summary_path)
 
         # 5) Send success email
         send_success_email(
@@ -318,3 +340,13 @@ def process_transcription_task(
             task_id=task_id,
         )
         raise e
+    finally:
+        # 6) Cleanup temporary files (best-effort)
+        for path in temp_files:
+            _remove_file(path)
+
+        # Also remove uploaded audio file
+        if audio_path:
+            _remove_file(audio_path)
+
+        logger.info("Cleanup completed for job %s.", task_id)