printing extracted text10

main.py

@@ -277,47 +277,40 @@ def new_value(value, old_min, old_max, new_min, new_max):
 @app.route('/compute_answers', methods=['POST'])
 def compute_answers():
     try:
-        query_file = request.files.get('query_file')
-        if not query_file:
-            error_msg = "Missing query file"
-            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
+        # For CSV files, process directly without any model initialization
         if file_type == "csv":
             ans_csv_file = request.files.get('ans_csv_file')
             if not ans_csv_file:
-                error_msg = "Missing answer CSV file"
-                log_print(error_msg, "ERROR")
-                return jsonify({"error": error_msg}), 400
+                return jsonify({"error": "Missing answer CSV file"}), 400
             
             try:
-                # 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()
+                # Read CSV content directly without unnecessary processing
+                content = ans_csv_file.read().decode('utf-8')
+                if not content.strip():
+                    return jsonify({"error": "CSV file is empty"}), 400
+                
+                # Process answers more efficiently
                 c_answers = []
-                for line in ans_csv_content:
-                    if line.strip():  # Skip empty lines
+                for line in content.splitlines():
+                    if line.strip():
                         answers = [ans.strip() for ans in line.split('\\n') if ans.strip()]
-                        c_answers.append(answers)
+                        if answers:  # Only add if there are valid answers
+                            c_answers.append(answers)
                 
                 if not c_answers:
-                    error_msg = "No valid answers found in CSV file"
-                    log_print(error_msg, "ERROR")
-                    return jsonify({"error": error_msg}), 400
+                    return jsonify({"error": "No valid answers found in CSV file"}), 400
                 
                 log_print(f"Successfully processed {len(c_answers)} answers from CSV")
                 return jsonify({"answers": c_answers}), 200
                 
             except Exception as e:
-                error_msg = f"Error processing CSV file: {str(e)}"
-                log_print(error_msg, "ERROR")
-                return jsonify({"error": error_msg}), 400
+                log_print(f"Error processing CSV file: {str(e)}", "ERROR")
+                return jsonify({"error": f"Error processing CSV file: {str(e)}"}), 400
         
-        # Only initialize models for PDF processing
+        # For PDF files, continue with existing processing
         elif file_type == 'pdf':
             # Wait for initialization only for PDF files
             wait_for_initialization()
@@ -345,7 +338,7 @@ def compute_answers():
                 database_creation(pdf_file)
             
             # Generate answers
-            queries = query_file.read().decode('utf-8').splitlines()
+            queries = request.files.get('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"
@@ -810,12 +803,15 @@ def wait_for_initialization():
     initialization_complete.wait()
     return True
 
-# Add this to the compute_marks route before processing files
 @app.before_request
 def ensure_initialization():
     """Ensure all resources are initialized before processing requests"""
-    if request.endpoint in ['compute_marks', 'compute_answers']:
+    if request.endpoint == 'compute_marks':
         wait_for_initialization()
+    elif request.endpoint == 'compute_answers':
+        # Only wait for initialization if processing PDF files
+        if request.form.get('file_type') == 'pdf':
+            wait_for_initialization()
 
 def cleanup_temp_files():
     """Clean up temporary files with proper error handling"""