import camelot
import pandas as pd
import re

def extract_bank_data(file_path):
    print(f"Reading {file_path}... enforcing strict Debit/Credit alignment.")
    
    # We use 'stream' flavor for the NLB PDF layout
    tables = camelot.read_pdf(file_path, pages='1-end', flavor='stream')
    
    final_transactions = []
    current_tx = None
    
    # Precise Regex Patterns
    iban_pattern = r'SI56\s?\d{4}\s?\d{4}\s?\d{4}\s?\d{3}'
    amount_pattern = r'=(\d+[\.,]\d{2})'
    date_pattern = r'(\d{2}\.\d{2}\.\d{2})'
    bic_pattern = r'\b[A-Z]{8,11}\b' # Standard BIC pattern

    for table in tables:
        df = table.df
        for _, row in df.iterrows():
            cells = [str(c).strip() for c in row]
            row_text = " ".join(cells)
            
            # TRIGGER: A new transaction starts with an amount (=xx,xx) and a date
            amount_match = re.search(amount_pattern, row_text)
            date_match = re.search(date_pattern, row_text)
            
            if amount_match and date_match:
                # Save previous transaction before starting new one
                if current_tx:
                    final_transactions.append(current_tx)
                
                amt_val = amount_match.group(1)
                
                # --- STRICT DEBIT/CREDIT LOGIC ---
                # Based on the NLB layout, columns are indexed 0 to N.
                # If the '=' sign is found in index 1, it is DEBIT.
                # If the '=' sign is found in index 2 or higher, it is CREDIT.
                is_credit = False
                for i, cell_content in enumerate(cells):
                    if '=' in cell_content:
                        if i >= 2: 
                            is_credit = True
                        break
                
                # Take only the first line of the cell as the Name (to exclude address)
                name_raw = cells[0].split('\n')[0].strip()

                current_tx = {
                    "value date": date_match.group(1),
                    "Creditor / Debtor": name_raw,
                    "account": "",
                    "payment details": "",
                    "debit": "" if is_credit else amt_val,
                    "credit": amt_val if is_credit else "",
                    "collecting": True
                }
                
                # Check if IBAN is on this same row
                ib = re.search(iban_pattern, cells[0])
                if ib:
                    current_tx["account"] = ib.group(0)

            elif current_tx:
                text_line = cells[0]
                if not text_line or text_line.lower() in ["address", "account"]:
                    continue

                # 1. Look for IBAN
                iban_search = re.search(iban_pattern, text_line)
                if iban_search:
                    current_tx["account"] = iban_search.group(0)
                    # Anything after IBAN on the same line might be details
                    rem = text_line.replace(iban_search.group(0), "").strip()
                    text_line = rem
                
                # 2. Filter out Address and Bank Noise
                is_address = any(k in text_line.upper() for k in ["CESTA", "ULICA", "TRG", " LJUBLJANA", " LOGATEC"]) or re.search(r'\d{4}', text_line)
                
                if not is_address and len(text_line) > 1:
                    # Clean up BIC/Routing codes (e.g. LJBASI2X)
                    clean_text = re.sub(bic_pattern, '', text_line).strip()
                    # Clean up internal bank references (CR.../DR...)
                    clean_text = re.sub(r'\b[CD]R\d+\b|\bNRC\b', '', clean_text).strip()
                    
                    if clean_text:
                        if not current_tx["payment details"]:
                            current_tx["payment details"] = clean_text
                        else:
                            # Avoid repeating the name in details if it's already there
                            if clean_text not in current_tx["Creditor / Debtor"]:
                                current_tx["payment details"] += " " + clean_text

    # Append last transaction
    if current_tx:
        final_transactions.append(current_tx)

    # Convert to DataFrame
    df_final = pd.DataFrame(final_transactions)
    
    # Final Cleanup: ensure columns match your requested list exactly
    cols = ["value date", "Creditor / Debtor", "account", "payment details", "debit", "credit"]
    df_final = df_final[cols]
    
    # Clean whitespace and handle empty values
    for col in df_final.columns:
        df_final[col] = df_final[col].astype(str).str.replace(r'\s+', ' ', regex=True).str.strip()
        df_final[col] = df_final[col].replace('nan', '')

    output_file = "bank_export_final_fixed.csv"
    df_final.to_csv(output_file, sep=';', index=False, encoding='utf-8-sig')
    print(f"Success! {len(df_final)} transactions exported to {output_file}")

if __name__ == "__main__":
    extract_bank_data("MSDataReport (1).pdf")