BR-CO-27 can not fail

[PaulKalbitzer]

As I understand it, the test for BR-CO-27 in its current form is a tautology.

<assert id=“BR-CO-27” flag=“fatal” test="(ram:PayeePartyCreditorFinancialAccount/ram:IBANID) or (ram:PayeePartyCreditorFinancialAccount/ram:ProprietaryID) or (not(ram: PayeePartyCreditorFinancialAccount/ram:IBANID) and not(ram:PayeePartyCreditorFinancialAccount/ram:ProprietaryID))">[BR-CO-27]- Either the IBAN or a Proprietary ID (BT-84) shall be used. </assert>

It is fulfilled if IBANID and ProprietaryID, only one of the two, or none at all are present

[phax]

Yep. It seems to be "a or b or (not (a) and not (b))".
It should be "(a or b) and not (a and b)" imho

[k-o-n-s-t]

I can confirm the error. The rule BR-CO-27 goes back to the Issue #132, proposing "a rule to check that tags IBANID and ProprietaryID are mutually exclusive should exist in CII schematron". Commit 9667fc4 closed that issue, unfortunately, with a wrong assert definition. In addition, BR-CO-27 rule never appeared in the EN16931 specification (DIN EN 16931-1 pdf).

[AndreasPvd]

This rule was intentionally created that way back at the first version. The reason was that it was the time of transition towards SEPA. In Germany, at that time, it was common to state the proprietaryID AND the IBAN at the same time for interoperability issues. That's why the implementation intentionally allowed to state both ath the same time and not to have them mutually exclusive. From a 2024 perspective I would suggest as well to change it to mutually exclusive - but I am not sure if it needs discussion in CEN in order to still support that kind option. Please do not forget that not all countries in the world are part of SEPA and in a B2B scenario proprietary IDs are still needed.