Netherlands national checksum validation

[PabloAJomer]

Hi, I'm curious where the original source for the national checksum validation for the Netherlands is?
I'm having trouble finding reliable sources but we have an IBAN that we have tried paying money to and it has worked, that is not valid according to Schwifty. Also I find some sources saying that there are two validation schemes for NL BBAN.

I don't find this source below that creditable but it seems to indicate that there are two validation algorithms of which one is the one implemented by the Schwifty.

https://www.credit-card.be/BankAccount/ValidationRules.htm#NL_Validation

Kindly advice.

[PabloAJomer]

I have some further proof. There used to be a so called "elfproef" rule which states to use the alg that is used. This may still be a valid check for some accounts but for newer account this rule seems to not be enforced.

Trying this IBAN: NL08BNPA8026000002 (which we have managed to pay to) we see that the below validator fails but it also states that it's normal to have the elfproof fail. So I suppose it becomes a bit sematic wthere it should be used or not? I haven't figured out if there is a way of knowing if it should be validated or not.

https://www.ibancalculator.com/iban_validieren.html

[Natim]

My version of schwifty doesn't raise for it:

>>> schwifty.IBAN("NL08BNPA8026000002").validate()
True

[Natim]

You are right, it's interesting

[Natim]

@sh4dowb since you implemented the validation rules for Netherlands, would you have more context about that?

[PabloAJomer]

I think it looks a lot like the elfproef algorithm in the code. elfproef basically means 11 test and as we can see in the code we multiply the 10 account digits with the vector and then compare to mod 11.

[PabloAJomer]

Also the fact that the bank issued us this iban where we could even make payments speaks in favour of something being wrong since it isn't accepted by the library.

[PabloAJomer]

It passes the IBAN validation but not when enabling BBAN validation just to be clear.

IBAN("NL08BNPA8026000002", validate_bban=True)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/Users/pablojomer/.pyenv/versions/3.11.5/envs/service-sepa-camt-handler/lib/python3.11/site-packages/schwifty/iban.py", line 77, in __init__
    self.validate(validate_bban)
  File "/Users/pablojomer/.pyenv/versions/3.11.5/envs/service-sepa-camt-handler/lib/python3.11/site-packages/schwifty/iban.py", line 180, in validate
    self.bban.validate_national_checksum()
  File "/Users/pablojomer/.pyenv/versions/3.11.5/envs/service-sepa-camt-handler/lib/python3.11/site-packages/schwifty/bban.py", line 145, in validate_national_checksum
    raise exceptions.InvalidBBANChecksum("Invalid national checksum")
schwifty.exceptions.InvalidBBANChecksum: Invalid national checksum

[mdomke]

@PabloAJomer 🤔 After reading https://www.betaalvereniging.nl/betaalproducten-en-diensten/iban/ibannext/ it seems to me that we should stop enforcing the checksum validation for Dutch IBANs, even though the validate_bban parameter has been set to True.

[PabloAJomer]

@mdomke Yes, I agree with that assessment. I don't know if there are cases where the "elfprooef" could still be useful but the administrative authority says that it should not be checked any more.