I think once we introduce the ValidatingWebhook in #100, then part of the .spec.models[] validation logic could live there. This is a place where more complex validation should reside, IMO, and it will nicely reject the creation of an OpenStackLightspeed CR instance with an invalid models section.
Originally posted by @lpiwowar in #107 (comment)
Note
This depends on #100
I think once we introduce the
ValidatingWebhookin #100, then part of the.spec.models[]validation logic could live there. This is a place where more complex validation should reside, IMO, and it will nicely reject the creation of an OpenStackLightspeed CR instance with an invalid models section.Originally posted by @lpiwowar in #107 (comment)