Update pipeline.py

pipeline.py

@@ -76,7 +76,10 @@ def find_drive_file(filename, parent_id):
 #     return file["id"]
 def upload_file_to_drive(local_path, remote_name, folder_id):
     try:
-        # Check if file already exists and delete it
+        if not os.path.exists(local_path):
+            raise FileNotFoundError(f"❌ Local file does not exist: {local_path}")
+
+        # Delete existing file on Drive if present
         existing = drive_service.files().list(
             q=f"name='{remote_name}' and '{folder_id}' in parents and trashed = false",
             fields="files(id)"
@@ -86,7 +89,6 @@ def upload_file_to_drive(local_path, remote_name, folder_id):
             drive_service.files().delete(fileId=existing[0]["id"]).execute()
             print(f"🗑️ Deleted existing '{remote_name}' in Drive folder {folder_id}")
 
-        # Now upload the file
         file_metadata = {"name": remote_name, "parents": [folder_id]}
         media = MediaFileUpload(local_path, resumable=True)
         file = drive_service.files().create(
@@ -94,19 +96,8 @@ def upload_file_to_drive(local_path, remote_name, folder_id):
             media_body=media,
             fields="id"
         ).execute()
-        print(f"✅ Uploaded '{remote_name}' to Google Drive folder ID: {folder_id}")
-
-        # Optional verification after upload
-        uploaded = drive_service.files().list(
-            q=f"name='{remote_name}' and '{folder_id}' in parents and trashed = false",
-            fields="files(id, name)"
-        ).execute().get("files", [])
-
-        if not uploaded:
-            print(f"❌ Upload verification failed: '{remote_name}' not found after upload.")
-        else:
-            print(f"✅ Verified upload: {remote_name} (ID: {uploaded[0]['id']})")
 
+        print(f"✅ Uploaded '{remote_name}' to Google Drive folder ID: {folder_id}")
         return file["id"]
 
     except Exception as e:
@@ -257,8 +248,10 @@ def pipeline_with_gemini(accessions):
           all_filename = f"{saveName}_all_merged_document.docx"
       print(chunk_filename, all_filename)  
       # Define local temp paths for reading/writing
-      import tempfile
-      tmp_dir = tempfile.mkdtemp()
+      # import tempfile
+      # tmp_dir = tempfile.mkdtemp()
+      tmp_dir = "/mnt/data/generated_docs"
+      os.makedirs(tmp_dir, exist_ok=True)
       file_chunk_path = os.path.join(tmp_dir, chunk_filename)
       file_all_path = os.path.join(tmp_dir, all_filename)
       # file_chunk_path = os.path.join(tempfile.gettempdir(), chunk_filename)