Update app.py

app.py

@@ -38,20 +38,14 @@ def extract_text_from_pdf(pdf_path, is_scanned=False):
 def parse_bank_statement(text):
     """Parse bank statement using LLM with fallback to rule-based parser"""
     cleaned_text = re.sub(r'[\x00-\x08\x0b\x0c\x0e-\x1f\x7f]', '', text)
-    safe_text = (
-        cleaned_text
-        .replace("{", "{{")
-        .replace("}", "}}")
-        .replace("%", "%%")
-    )
     print(f"Original text sample: {cleaned_text[:200]}...")
-    print(f"Safe text sample: {safe_text[:200]}...")
     
-    # Craft precise prompt for LLM
+    # Craft precise prompt for LLM with proper JSON escaping
     prompt = f"""
 You are a financial data parser. Extract transactions from bank statements.
 
 Given this bank statement text:
+{cleaned_text}
 
 Extract all transactions with these fields:
 - Date
@@ -65,10 +59,28 @@ Extract all transactions with these fields:
 Return JSON with "transactions" array containing these fields.
 
 Example format:
-{"transactions": [
-  {"date": "2025-05-08", "description": "Company XYZ Payroll", "amount": "8315.40", "debit": "0.00", "credit": "8315.40", "closing_balance": "38315.40", "category": "Salary"},
-  ...
-]}
+{{
+  "transactions": [
+    {{
+      "date": "2025-05-08",
+      "description": "Company XYZ Payroll",
+      "amount": "8315.40",
+      "debit": "0.00",
+      "credit": "8315.40",
+      "closing_balance": "38315.40",
+      "category": "Salary"
+    }},
+    {{
+      "date": "2025-05-19",
+      "description": "Whole Foods",
+      "amount": "142.21",
+      "debit": "142.21",
+      "credit": "0.00",
+      "closing_balance": "38173.19",
+      "category": "Groceries"
+    }}
+  ]
+}}
 
 Rules:
 1. Ensure numeric fields have valid numbers (e.g., "0.00" instead of "-")