You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
All responses in ResponseMessageGroup [1] require the MessageBlockID field to be present. The thinking behind this is to make processing mainly for legacy systems easier, which cannot cope with huge message sizes. Since the field is also not optional, this has some consequences.
First, we always must set it to some dummy value, just to pass validation and make the recipient's parser happy. Secondly, a bizarre side effect is that even though there was a problem generating a response, and we send an IXSI error object back, this MessageBlockID field must still be included [2], which doesn't make sense. As it turns out, the schema needs some refinements in this area.
All responses in ResponseMessageGroup [1] require the MessageBlockID field to be present. The thinking behind this is to make processing mainly for legacy systems easier, which cannot cope with huge message sizes. Since the field is also not optional, this has some consequences.
First, we always must set it to some dummy value, just to pass validation and make the recipient's parser happy. Secondly, a bizarre side effect is that even though there was a problem generating a response, and we send an IXSI error object back, this MessageBlockID field must still be included [2], which doesn't make sense. As it turns out, the schema needs some refinements in this area.
[1] https://github.com/RWTH-i5-IDSG/ixsi/blob/master/xml/IXSI.xsd#L1896
[2] https://github.com/RWTH-i5-IDSG/BikeMan/blob/sessionbased-auth/src/main/java/de/rwth/idsg/bikeman/ixsi/processor/subscription/complete/CompleteAvailabilityRequestProcessor.java#L60-L62
The text was updated successfully, but these errors were encountered: