Support checking for invalid CSV files for DKB (Germany)

[denkristoffer]

👋

DKB in Germany exports invalid CSV files that need a bit of manipulation to work with this module. Specifically, the first 6 lines of the file need to be removed. This would have to happen before the current code would be able to identify the file as DKB data. Is this a use case you are interested in allowing for?

Here's an example export to show you the problem

"Kontonummer:";"DE11111111111111111111 / Konto";

"Von:";"23.10.2021";
"Bis:";"22.11.2021";
"Kontostand vom 22.11.2021:";"1.111,11 EUR";

"Buchungstag";"Wertstellung";"Buchungstext";"Auftraggeber / Beg�nstigter";"Verwendungszweck";"Kontonummer";"BLZ";"Betrag (EUR)";"Gl�ubiger-ID";"Mandatsreferenz";"Kundenreferenz";
"27.10.2021";"27.10.2021";"Lohn";"Acme";"Lohn";"DE11111111111111111111";"ABCDEFGH111";"1.111,11";"";"";"111111111/111111";

[jorisnoo]

Hi @denkristoffer

Thanks for your message and your interest. I view this as a tool to allow imports from files the built-in importer struggles with — so yes, this is a case that ideally would be supported. You're right, however, that changes to the current logic would be required to make this work.

If you have an idea how it could be solved, please feel free to make a PR! Otherwise I'll have a look at this in January, as soon as my schedule allows.

[denkristoffer]

@jorisnoo

What would you say to moving the bank matching logic to the bank class and then letting each bank define its own way of matching? One challenge I can see is that DKB needs to be matched before parsing the CSV while all the existing banks would need to be matched after parsing.