CODEWITHSUNDEEP
python-3.xregexlistdictionarynamedtuple
Subhojyoti Lahiri
Subhojyoti Lahiri

Reputation: 111

Regex from a list of strings in Python

I have a list named as Statement created from pdf using pytesseract and Regex:

Statement= ['07-10-2019 UPI/927912685773/UPI/surya.balaji94@/Citibank 6,677.00 2,36,804.08',
         '07-10-2019 MOBILE BANKING DUT/CITIBANK 3,403.00 2,40,207.08',
         '07-10-2019 BIL/INFT/001818195728/82D3/ AJAY KUMAR JHA 6,080.00 2,46, 287.08',
         '08.10.2019 MOBILE BANKING MMM TiMPS/928115182374/8161 Oct Mte/AMARJEET SIHDFC 4,411.00 250,698.08',
         '08-10-2019 BIL/INFT/001818636132/E3 BIk1 Pramod/ PRAMOD KUMAR P 6,599.00 2,57,297.08']

With some help on stack, I created a list of dictionaries as follows:

cols = ["Date", "Item_Name", "Transaction_Amount", "Balance"]
date_pattern = re.compile(r"\d{2}[- /.]\d{2}[- /.]\d{4}", re.I)
item_and_name_pattern = re.compile(r"(?<=\d{2}-\d{2}-\d{4}\s).*", re.I)
amount_pattern = re.compile(r"\d+,\d+.\d+", re.I)
total_pattern = re.compile(r"\d+,\d+,\d+.\d+$", re.I)
Transaction = namedtuple("Transaction", cols)
transactions = []
for item in Statement:
    try:
        date = re.search(date_pattern, item).group()
        total = re.search(total_pattern, item).group()
        temp_1 = item.rstrip(total)
        amount = re.search(amount_pattern, item).group()
        temp_2 = temp_1.strip().rstrip(amount)
        item_and_name = re.search(item_and_name_pattern, temp_2).group()
    except:
        pass
    t = Transaction(date, item_and_name, amount, total)
    transactions.append(t)
    out = [{k:v for k, v in f._asdict().items()} for f in transactions]

But the output is not satisfactory as it is picking up the date but going wrong with item name and total etc for that date(check the list above and match with the dictionaries below). I want to know if there is any other way to store them in named columns correctly?

    [{'Date': '07-10-2019',
  'Item_Name': 'UPI/927912685773/UPI/surya.balaji94@/Citibank ',
  'Transaction_Amount': '6,677.00',
  'Balance': '2,36,804.08'},
 {'Date': '07-10-2019',
  'Item_Name': 'MOBILE BANKING DUT/CITIBANK ',
  'Transaction_Amount': '3,403.00',
  'Balance': '2,40,207.08'},
 {'Date': '07-10-2019',
  'Item_Name': 'MOBILE BANKING DUT/CITIBANK ',
  'Transaction_Amount': '3,403.00',
  'Balance': '2,40,207.08'},
 {'Date': '08.10.2019',
  'Item_Name': 'MOBILE BANKING DUT/CITIBANK ',
  'Transaction_Amount': '3,403.00',
  'Balance': '2,40,207.08'},
 {'Date': '08-10-2019',
  'Item_Name': 'BIL/INFT/001818636132/E3 BIk1 Pramod/ PRAMOD KUMAR P ',
  'Transaction_Amount': '6,599.00',
  'Balance': '2,57,297.08'}]

Upvotes: 1

Views: 167

Answers (1)

Ronie Martinez
Ronie Martinez

Reputation: 1266

Here is a simpler way:

import re

pattern = re.compile("(?P<Date>\d{2}[.-]\d{2}[.-]\d{4})\s(?P<Item_Name>.+)\s(?P<Transaction_Amount>[0-9,\.]+)\s(?P<Balance>[0-9,\.]+)")

print([pattern.match(item).groupdict() for item in Statement])

EDIT: If using try-except as requested in the comments:

result = []
for item in Statement:
    try:
        result.append(pattern.match(item).groupdict())
    except AttributeError:
          pass

print(result)

Upvotes: 1

Related Questions