Invoice processing

import re
import pandas as pd
import tkinter as tk
from tkinter import filedialog
import datetime
from datetime import date
import pdfplumber

today = str(date.today())

#the dataframe that gets exported to excel, columns will show up in row 1
exceldf = pd.DataFrame(columns=['Invoice Number', 'Bill or Credit', 'Invoice Date', 'Accounting Date', 'Vendor', 'Invoice Description', 'Line Item Number', 'Property Abbreviation', 'Location', 'GL Account', 'Line Item Description', 'Amount', 'Expense Type', 'Is Replacement Reserve', 'Payment Date', 'Reference Number', 'Payment Method', 'Department', 'Due Date', 'Quantity', 'Unit Price', 'Tax', 'Received Date'])

text = ""
x = 1
y = 1

#Functions to process the text pulled from the pdf
def find_invoice_numbers(text):
    #Finds all instances of INVOICE IXXXXXXXXX
    matches = re.findall(r'^I\d{10}', text, flags=re.MULTILINE)
    return list(set(matches))

def find_account_numbers(text):
    #Finds account number, starts with A and goes 10 digits
    total_pattern = r'A\d{10}'
    matches = re.findall(total_pattern, text, re.MULTILINE)
    return list(dict.fromkeys(matches))

def find_amount_due(text):
    #Finds all numbers with a decimal point. Will just use the first one found in the je below
    total_pattern = r'\d*\,?\d*\.\d{2}'
    matches = re.findall(total_pattern, text, re.MULTILINE)
    return list(dict.fromkeys(matches))

def find_invoice_dates(text):
    #Finds all dates and returns no duplicates, will just use the first one on each page
    total_pattern = r'\d{2}\/\d{2}\/\d{4}'
    matches = re.findall(total_pattern, text)
    return list(dict.fromkeys(matches))

def get_prop_abbr(account_no):
    #uses the account number return property code
    match account_no:
            case 'acct1': return 'prop1'
            case 'acct2': return 'prop2'
            case 'acct3': return 'prop3'
            case 'acct4': return 'prop4'
            case _: return ''

#opens dialogue box to select the pdf
pdf_document = filedialog.askopenfilename()
pdf= pdfplumber.open(pdf_document)

#writes to dataframe that will then be written to excel
for page in pdf.pages:
    text = page.extract_text()

    #passes through text to the various functions above
    invoicenum = find_invoice_numbers(text)
    account_num = find_account_numbers(text)
    amount_due = find_amount_due(text)
    invoice_dates = find_invoice_dates(text)

    #using only first result for invoicenum and account_num
    exceldf.loc[y, 'Invoice Number'] = invoicenum[0]
    exceldf.loc[y, 'Bill or Credit'] = 'Bill'
    exceldf.loc[y, 'GL Account'] = 69750
    exceldf.loc[y, 'Invoice Description'] = 'monthly service'
    exceldf.loc[y, 'Line Item Description'] = 'monthly service'
    exceldf.loc[y, 'Invoice Date'] = invoice_dates[0]
    exceldf.loc[y, 'Accounting Date'] = date.today().strftime("%m/%d/%Y")
    exceldf.loc[y, 'Vendor'] = 'VEND1'
    exceldf.loc[y, 'Line Item Number'] = 1
    exceldf.loc[y, 'Property Abbreviation'] = get_prop_abbr(account_num[0])

    #removes any comma and converts it into a float (so it's not a literal)
    exceldf.loc[y, 'Amount'] = float(amount_due[0].replace(",",""))
    exceldf.loc[y, 'Expense Type'] = 'General'
    exceldf.loc[y, 'Is Replacement Reserve'] = 'FALSE'
    y+= 1


pdf.close()
#saves excel file to invoiceuploads folder

exceldf.to_excel('invoiceuploads/vendorupload' + today + '.xlsx', index=False)
print('A file is now ready to be uploaded in /invoiceuploads.')