kamelet: allow to add additional attributes to in/out

[lburgazzoli]

As today, a kamelet has the option to define the types it supports but that is limited to the following information:

type EventTypeSpec struct {
	// media type as expected for HTTP media types (ie, application/json)
	MediaType string `json:"mediaType,omitempty"`
	// the expected schema for the event
	Schema *JSONSchemaProps `json:"schema,omitempty"`
}

Given that a kamelet may be more complex and the tooling using them may need more information it would be nice to have the option to add additional metadata, as example, in some case it may be useful to add an information about what is the Java class the output of a camel is build upon:

spec:
  out:
    mediaType: "accplication/json"
    annotations:
        contentClass: com.acme.Foo

This may be useful for tools as they may compute the scheme of the payload.

[github-actions]

This issue has been automatically marked as stale due to 90 days of inactivity.
It will be closed if no further activity occurs within 15 days.
If you think that’s incorrect or the issue should never stale, please simply write any comment.
Thanks for your contributions!

[squakez]

Given #4100 - is this still relevant @christophd ?

[christophd]

I think partially this has been done already with #4100

The Java contentClass class information that is explicitly mentioned in this issue is missing though in the DataTypes section in Kamelet spec AFAIK.

Information such as the Java contentClass may be given on the individual schema resolver step using the resolve-pojo-schema-action Kamelet. The idea is to resolve the schema for a contentClass (delegating to Avro, Protobuf, ...) in a first step and the have the information available to serialize/deserialize the content using one of the serdes action Kamelets (e.g. avro-deserialize-action

@lburgazzoli WDYT?

[lburgazzoli]

If I properly recall the problem, here the idea was more related to design time, so to provide hints for tooling i.e. if you have a data mapper then you can use the schema in the mapper. But I think we can close this issue and eventually open a new and more detailed one.