# Importing libraries
from numpy import genfromtxt, savetxt
from time import time
from datetime import datetime
import models
import os, re
from database import engine, SessionLocal
from pytz import timezone

def Load_Data(file_name):
    #data = genfromtxt(file_name, delimiter=';', skip_header=1, converters={0: lambda s: str(s)})
    #str2date = lambda x: datetime.strptime(x, '%d-%m-%y')
    data = genfromtxt(file_name, delimiter=';', dtype=None, skip_header=0, encoding='utf-8-sig')

    return data.tolist()

def extract_values(text):
    # Regular expressions to match the required fields
    regex_patterns = {
        'Auftraggeber': r'Auftraggeber: (.*?) (Verwendungszweck|Zahlungsreferenz|IBAN|BIC)',
        'Verwendungszweck': r'Verwendungszweck: (.*?) (Zahlungsreferenz|IBAN|BIC)',
        'Zahlungsreferenz': r'Zahlungsreferenz: (.*?) (Verwendungszweck|IBAN|BIC)',
        'IBAN': r'IBAN Auftraggeber: (.*?) ',
        'BIC': r'BIC Auftraggeber: (.*?)$'
    }
    
    # Initialize an empty list to store extracted information
    extracted_info = []
    
    # Initialize an empty dictionary to store information from this row
    row_info = {}
    
    # Extract information using regular expressions
    for key, pattern in regex_patterns.items():
        match = re.search(pattern, text)
        if match:
            row_info[key] = match.group(1)
        else:
            row_info[key] = None
    
    # Append the extracted information from this row to the list
    extracted_info.append(row_info)
    
    return extracted_info

if __name__ == "__main__":
    t = time()

    db = SessionLocal()
    models.Base.metadata.create_all(bind=engine)

    try:
        home_directory = os.path.expanduser('~')
        source_csv = os.path.join(home_directory, "OneDrive", "Dokumente", "SUST", "UmsatzListe_AT493443700000113407_20241212_095142_mod.csv")
        target_csv = source_csv.replace(".csv", "_out.csv")
        data = Load_Data(source_csv) 

        tz = timezone('UTC')  # Replace with your desired timezone
        outData = []
        for i in data:
            if ('Auftraggeber' in i[1]):
                # extract key-value pairs Auftraggeber, Verwendungszweck, Zahlungsreferenz, IBAN
                new_list = extract_values(i[1])

                if len(new_list) > 0:
                    outData.append([i[0], new_list[0].get('Auftraggeber'), new_list[0].get('Zahlungsreferenz'), new_list[0].get('Verwendungszweck'), new_list[0].get('IBAN'), i[2], i[3].replace(',','.'), i[4], i[5]])
        
        savetxt(target_csv, outData, delimiter=";", encoding='utf-8', header='valuta_date;client;reference;intended;IBAN;entry_date;amount;currency;timestamp', fmt='%s')

    except Exception as e:
        print (e)
    finally:
        print("Time elapsed: " + str(time() - t) + " s.") #0.091s