Changes for week of July 7#384
Conversation
- more minor edits from reviews Signed-off-by: Doug Davis <duglin@gmail.com>
- ximportresources are now transitive - more typos Signed-off-by: Doug Davis <duglin@gmail.com>
Signed-off-by: Doug Davis <duglin@gmail.com>
| the length of the word `count`. | ||
| Attribute names and key names are limited to 63 characters, so why are some | ||
| Group and Resource names limited to less? Because when they appear as part of | ||
| attribute names and they can be append with phrase like `url`, `count` |
| shared compatible model. | ||
| - Does not imply runtime validation of the claim. | ||
| - Example: `https://raw.githubusercontent.com/xregistry/spec/refs/heads/main/schema/model.json` | ||
| - Example: `http://xregistry.io/xreg/xregistryspecs/schema-v1/docs/model.json` |
There was a problem hiding this comment.
Is this deliberately http now instead of https?
There was a problem hiding this comment.
nope - made it https again
| combination defined within the same Registry. It MUST NOT reference the | ||
| same Group under which the `ximportresources` resides. | ||
| - An empty array MAY be specified, implying no Resources are imported. | ||
| - The use of `ximportresources` MAY be transitive as long as it does not |
There was a problem hiding this comment.
This "MAY" feels ambiguous to me. Are we saying that the server may support transitive imports but doesn't have to, or that a particular usage may be transitive but it's valid to import a resource that doesn't import anything itself? If it's the latter, I think it would be better to say "The use of ximportresources MUST be handled in a transitive fashion [...]" or similar - to make it clear that this is really a requirement of the server.
There was a problem hiding this comment.
Is this any better?
- The definition of a Group MAY include an `ximportresources` directive
that references a Resource from another Group that itself is defined
via an `ximportresources`. However, transitive definitions of Resources
MUST NOT result in a circular import chain.
There was a problem hiding this comment.
Yes, I think so. Hard to say exactly why, but it feels like it's leaving less wiggle room for servers.
There was a problem hiding this comment.
(Although maybe we should say that a server MUST spot circular import chains and report an error in that case.)
There was a problem hiding this comment.
I did think about that but decided against it because we don't go out of our way to say that for all "the client MAY" type of statements. I think it should be obvious.
Signed-off-by: Doug Davis <duglin@gmail.com>
|
Approved on 7/10 call |
2 participants
commit 1
commit 2
commit 3
commit 4
From last week: