fidorsmart Import

[Martin Steigerwald]

Hello!

I start this with an recommendation to all finance institutes:

Do not invent your own format! Simply just don´t.

Or isn´t there some standard at all?

Yes, there is HBCI/FinTS, but they chose to use their own API. But for CSV 
export? It is always *the same* kind of data. Why have a ton of different CSV 
formats to export them?


On the good side I can get a CSV export with fidorsmart.

I can also import it.

It has the following fields:

Datum;Beschreibung;Beschreibung2;Wert

15.08.2016;MasterCard Transaktion in Höhe von 9,99 EUR bei 
COMPANY.COM;"";-9,99

Additionally it isn´t UTF-8 but ISO-8859 text.


Which leads to the following issues:

1. Umlauts are missing

2. If I set field 2 as the payee I get the whole text of that in that field, 
but I prefer to have this in the description field.

3. If I set field 2+3 as description, KMyMoney doesn´t let me continue, as I 
don´t specify a payee field.


I´d say the way of least resistance might be to provide two enchancement 
requests for KMyMoney CSV import, one for the encoding, and one for the 
description thing. It may take a lot more effort to convince Fidor to 
introduce some sanity in their export format or even just support HBCI/FinTS.

Thanks,
-- 
Martin