Credentials have different structures at the policy level depending on the format

[u-veys]

I tried to use the same credential, once specified as json-ld and once as jwt. I have noticed that the same credential can have different structures at policy-level depending on whether it is jwt or ldp-credential. The reason for this appears to be the expansion of the credental.

While the json-ld credential is expanded, this is not the case with the jwt. On the one hand, I could not find any place in the code where the jwt is expanded, and on the other hand, I get a jwt-credential in the policy function that is not expanded. I can recognize this by the maps that 1) do not contain the context in the key and 2) do not have the structure that an expanded credential would have. However, the same credential, only specified as json-ld, is an expanded-credential in the policy function. This results in the following difference in the policy function for the same credential, only specified in a different formats (example shows only one claim of the credential):

ldp-credential:

"http://example.com/1.0/name": [
      { "@value": "Test" }
    ]

jwt-credential:

"name": "Test"

It therefore follows that the credential must be resolved differently in the policy function depending on the format, especially because of the different structure. This raises the question of whether this is what is intended and if so, how should/can it best be handled at policy level?