search

ietf-rats / ietf-corim-cddl

This repository is abandoned. The adopted I-D can be found at:
https://github.com/ietf-rats-wg/draft-ietf-rats-corim/
2 stars 0 forks source link

why is `element-name-map` mandatory in `endorsed-value-map` ? #33

Closed thomas-fossati closed 3 years ago

thomas-fossati commented 3 years ago

I can't recall why in endorsed-value-map we made element-name-map mandatory?

I.e., could we have instead:

endorsed-value-map = non-empty<{
  ? comid.element-name => element-name-map
  ? comid.element-value => element-value-map
  * $$endorsed-value-map-extension
}>
nedmsmith commented 3 years ago

maybe it is historical. The element-name is the 'key' in key-value pair. So, it plays an important role in matching claims. The Endorsement spec section 6 now has an improved algorithm section that helps describe matching. The schema can be more flexible in terms of optionality, but the normative might be tighten it up.

thomas-fossati commented 3 years ago

We have a couple of cases where it is redundant - as it just replicates the "outer" module name - which may benefit from elision.

(Besides, this seems to also apply to the reference-claim-map.)

nedmsmith commented 3 years ago

So the semantics would be that if there is a claim without an element-name it defaults to the module-name context?

thomas-fossati commented 3 years ago

So the semantics would be that if there is a claim without an element-name it defaults to the module-name context?

yes