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
I'm implementing DCQL right now (draft 24) and it's not clear to me from the spec how to implement value matching. Some observations
The standard says about values that it's "An array of strings, integers or boolean values that specifies the expected values of the claim." ... however it goes together with a path so the value to compare against might indeed be an object (e.g. "path": ["address"]) where address is an object with street_address, postal_code etc elements. Why don't we allow values to also include objects?
For ISO mdoc, it's not at all clear how to match. Yes, trivial cases where the data element is a tstr or boolean (e.g. major type 7) or integers (major type 0 and 1)... but what about things like full-date (which according to RFC 8943 is a string with tag 1004), for example, should we automatically drop all tags? In that case you can use e.g. values = ["2025-02-11"] to match a full-date. I think we should have concise language for the implementer here and not have them guess what the intent is.
The text was updated successfully, but these errors were encountered:
I'm implementing DCQL right now (draft 24) and it's not clear to me from the spec how to implement value matching. Some observations
The standard says about
values
that it's "An array of strings, integers or boolean values that specifies the expected values of the claim." ... however it goes together with a path so the value to compare against might indeed be an object (e.g."path": ["address"]
) whereaddress
is an object withstreet_address
,postal_code
etc elements. Why don't we allowvalues
to also include objects?For ISO mdoc, it's not at all clear how to match. Yes, trivial cases where the data element is a
tstr
or boolean (e.g. major type 7) or integers (major type 0 and 1)... but what about things likefull-date
(which according to RFC 8943 is a string with tag 1004), for example, should we automatically drop all tags? In that case you can use e.g.values = ["2025-02-11"]
to match afull-date
. I think we should have concise language for the implementer here and not have them guess what the intent is.The text was updated successfully, but these errors were encountered: