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
An interesting discussion started over here: yannh/kubeconform#171
A user is pointing out (correctly I believe) that we don't really make the distinction for strict vs non-strict validation for CRDs, whereas kubernetes-json-schema has schemas for both "strict" and "non-strict" mode. The strict schemas will complain when keys are used in the resource that are not present in the schema, vs the non-strict mode will not count this as an error.
In the user's case, they use a self-generated json schema, but I was wondering what the viewpoint was here - whether CRDs-catalog should also provide a strict and non-strict JSON schema for each CRD, or what the default really should be (I would be tempted to say, default to strict?).
Best,
Yann
The text was updated successfully, but these errors were encountered:
To be frank, we didn't consider this when we started this project.
we are using the schema that is generated by default via the openapi2jsonschema.py script, and it looks like those are not strict.
I don't see any reason why not to address this issue, so if the community would like to submit PRs we will accept them :)
An interesting discussion started over here: yannh/kubeconform#171
A user is pointing out (correctly I believe) that we don't really make the distinction for strict vs non-strict validation for CRDs, whereas kubernetes-json-schema has schemas for both "strict" and "non-strict" mode. The strict schemas will complain when keys are used in the resource that are not present in the schema, vs the non-strict mode will not count this as an error.
In the user's case, they use a self-generated json schema, but I was wondering what the viewpoint was here - whether CRDs-catalog should also provide a strict and non-strict JSON schema for each CRD, or what the default really should be (I would be tempted to say, default to strict?).
Best,
Yann
The text was updated successfully, but these errors were encountered: