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
But there are several problems with pointing folks to this technique:
It's not mentioned in the OAS spec, so many people, including code generation tool developers, are not aware of it
It's not even mentioned in the JSON Schema spec as it wasn't how I was thinking about it when I wrote those sections
$dynamicAnchor and $dynamicRef are among the most likely features to be hiding in the asterisk on many tools' "Supports OAS 3.1*" claims (it's basically taken discriminator's place as the keyword that is left until the end to even attempt)
We already have a large section on supporting polymorphic types. We should add a brief section on generic types, at least enough to make it clear what keywords can be used. We should definitely do this in 3.2.0, and maybe even 3.1.1 as it does not add any new requirements. In 3.2.0, I suppose we could add requirements to support the idiom but I'd settle for making people aware of it. If there's demand for it, tools will add support, but first folks have to know about it.
I don't think we need to go into a lot of detail on how it works. That can be offloaded to the Learn site.
The text was updated successfully, but these errors were encountered:
During the recent issue backlog tidying effort, I closed at least six issues by pointing to @gregsdennis's Using Dynamic References to Support Generic Types blog post:
But there are several problems with pointing folks to this technique:
$dynamicAnchor
and$dynamicRef
are among the most likely features to be hiding in the asterisk on many tools' "Supports OAS 3.1*" claims (it's basically takendiscriminator
's place as the keyword that is left until the end to even attempt)We already have a large section on supporting polymorphic types. We should add a brief section on generic types, at least enough to make it clear what keywords can be used. We should definitely do this in 3.2.0, and maybe even 3.1.1 as it does not add any new requirements. In 3.2.0, I suppose we could add requirements to support the idiom but I'd settle for making people aware of it. If there's demand for it, tools will add support, but first folks have to know about it.
I don't think we need to go into a lot of detail on how it works. That can be offloaded to the Learn site.
The text was updated successfully, but these errors were encountered: