yves-bussard
commented
4 years ago The link to the connector example. AccountSpiImpl calls the ledger with the resourceId: https://github.com/adorsys/xs2a-connector-examples/blob/develop/xs2a-connector/src/main/java/de/adorsys/aspsp/xs2a/connector/spi/impl/AccountSpiImpl.java
ViraHavrylenko
Place where bug appeared
Current behavior
Expected behavior
Steps to reproduce
When the TPP calls the resource /v1/accounts/{account-id} then we end up in AccountSpiImpl.requestAccountDetailForAccount(). The argument SpiAccountReference is created in the mapper Xs2aToSpiAccountReferenceMapper and then passed along to the method AccountSpiImpl.requestAccountDetailForAccount(). Unfortunately, the mapper creates the SpiAccountReference using the constructor WITHOUT the aspspId argument. Therefore this field is never set.
SCA approach
XS2A version(s):
Log files or other additional info
My understanding of the documentation is that the TPP never knows about the aspspId and the banking backend never knows about the resourceId. Therefore it should be the XS2A Server which maps a resourceId to an aspspId and the other way around. Since the framework does not pass along the aspspId from the AccountReference to the SpiAccountReference how can I get to the aspspId in the AccountSpiImpl? And if it is not supposed to be set in the SpiAccountReference, why is there even this field?
The only way I can make this work now is to set the resourceId and aspspId to the same value (the id from the backend) and use whatever field is not null. But that's probably not the idea.
BTW in the example connectors I saw that you call the ledger using the resourceId. Shouldn't you call the ledger (the banking backend) with the aspspId?