Drop rest-client template ?

[iocanel]

I think that now that we have a more concrete yet generic enough example (the chatbot) that shows how to consume an api, maybe we should drop the rest-client template.

[cmoulliard]

As the goal of the rest-client template is to:

• Generate a quarkus application using the extensions

• io.quarkus:quarkus-rest-jackson
• io.quarkus:quarkus-smallrye-openapi
• io.quarkus:quarkus-smallrye-graphql
• io.quarkus:quarkus-hibernate-orm-rest-data-panache

• Add some maven

        dependencies:
          - groupId: io.quarkiverse.openapi.generator
            artifactId: quarkus-openapi-generator
            version: 2.4.2
          - groupId: io.quarkus
            artifactId: quarkus-rest-client-jackson

  but without the skeleton of an application AND using only the quarkus starters then it don't really bring more added value

[cmoulliard]

I think that it is time to think about how we design a new Quarkus Application template.

Should we copy/paste an existing one and change the code with the risk that we will duplicate many common resources: Tekton, ArgoCD, Helm, etc, increase the maintenance cost when changes should be propagated to x template folders, etc

or instead

Should we try to create a generic Quarkus Application template able to deal with different examples such as ChatBot, TODO, REST server/client, etc and where the examples are handled under their specific skeleton folder but selected using an enum. Here is an example where the CI skeletons are selected using enum, etc - 

@iocanel @aureamunoz

[aureamunoz]

It depends on if we have deadlines that might favor a quicker solution or how often we will need to do changes on the template but in general I think the second solution is the better long-term solution despite its initial complexity. This approach makes maintenance easier because centralizes common functionality and can reduce duplication than copy/pasting and dealing with existing code.