printing extracted text9

main.py

@@ -283,23 +283,10 @@ def compute_answers():
             log_print(error_msg, "ERROR")
             return jsonify({"error": error_msg}), 400
         
-        # Read and validate query file
-        try:
-            queries = query_file.read().decode('utf-8').splitlines()
-            queries = [q.strip() for q in queries if q.strip()]
-            if not queries:
-                error_msg = "Query file is empty"
-                log_print(error_msg, "ERROR")
-                return jsonify({"error": error_msg}), 400
-            log_print(f"Successfully read {len(queries)} queries")
-        except Exception as e:
-            error_msg = f"Error reading query file: {str(e)}"
-            log_print(error_msg, "ERROR")
-            return jsonify({"error": error_msg}), 400
-        
         file_type = request.form.get('file_type')
         log_print(f"Processing file type: {file_type}")
         
+        # For CSV files, skip model initialization
         if file_type == "csv":
             ans_csv_file = request.files.get('ans_csv_file')
             if not ans_csv_file:
@@ -308,7 +295,7 @@ def compute_answers():
                 return jsonify({"error": error_msg}), 400
             
             try:
-                # Read and process CSV answers
+                # Read and process CSV answers directly without loading models
                 log_print("Reading CSV answer file...")
                 ans_csv_content = ans_csv_file.read().decode('utf-8').splitlines()
                 c_answers = []
@@ -330,7 +317,11 @@ def compute_answers():
                 log_print(error_msg, "ERROR")
                 return jsonify({"error": error_msg}), 400
         
+        # Only initialize models for PDF processing
         elif file_type == 'pdf':
+            # Wait for initialization only for PDF files
+            wait_for_initialization()
+            
             # Create a temporary directory for PDF files
             pdf_dir = tempfile.mkdtemp()
             os.makedirs(pdf_dir, exist_ok=True)
@@ -354,6 +345,14 @@ def compute_answers():
                 database_creation(pdf_file)
             
             # Generate answers
+            queries = query_file.read().decode('utf-8').splitlines()
+            queries = [q.strip() for q in queries if q.strip()]
+            if not queries:
+                error_msg = "Query file is empty"
+                log_print(error_msg, "ERROR")
+                return jsonify({"error": error_msg}), 400
+            
+            c_answers = []
             for query in queries:
                 ans = []
                 for pdf_file in pdf_files: