Skip to content

CarliJoy/GLS_account_statement_reader

Repository files navigation

Description

This script allows to read PDF bank statements from the GLS Bank. Even so, the banks online manager supports exporting of CSV files, this works only for the past three month whereas PDF account statements are kept for at least two years.

So if, you forced to automatically analyse past bank transactions, this script will help you.

The tool also supports reading the CSV files (which include more information), so you can analyse them.

I also tested it with Banking records from the Triodos Bank and it works well. They both using the same banking system, so maybe also other "Volksbank" or "Raiffeisenbank" work as well. Write me an issue

I tested it with the following Banking records so far:

  • GLS 2014-2020
  • Triodos 2020

Dependencies

  • python >= 3.6
  • pdftotext(part of poppler-utils)
  • jupyter-notebook [Optional]

Installation

Variant One

Install using pip install bank-statement-reader

Advantage:

  • Easy to install

Disadvantage:

  • some features like the booking/personal.py file depend on modifying the package the package source before installing, which won't work using this method

Variant Two

Clone the repro locally, create and activate a new virtual environment and run pip install -e . within the project folder.

Usage

After installation, you have a new command statement2csv available.

usage: statement2csv [-h] [--out out.csv] statement.pdf [statement.pdf ...]

Convert banking statements (PDF & CSV) to an analysed standard csv form.

positional arguments:
  statement.pdf  files to open and convert

optional arguments:
  -h, --help     show this help message and exit
  --out out.csv  csv file to write the results to

        If no filename is given, the file will be saved to
            basename_first_file_%date_string%.csv.
        %date_string% will be always replaced to 'YYYY-mm-dd_to_YYYY-mm-dd'
                                                 start date  to   end date

Another way to use the project is to use jupyter-notebook for fast analysing data. See example.ipynb for an idea how to use it.

Data Protection Note

As bank statement data is highly sensitive, only very general rules for categorizing were pushed to this git.

Use src/bank_statement_reader/bookings/personal.py for customizations of categories and payees. You only to create this file with a content like the following, and it will be used automatically.

from bank_statement_reader.booking.booking_base import BookingBase

class Booking(BookingBase):
    def _set_payee(self, value: str):
        """your custom functions here"""
        super()._set_payee(value)

    def _get_category(self):
        """your custom stuff here"""
        return super()._get_category()