[
https://issues.apache.org/jira/browse/SLING-9903?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17230679#comment-17230679
]
Henry Kuijpers commented on SLING-9903:
---------------------------------------
I agree with this. I already thought it was a bit of overkill to start a sling
instance for it etc etc, if we can also just validate the results of those
annotations by looking at what they produced.
I agree that the scope of the tests for this module should be limited to just
that. Maybe the IT that was created better suits the Sling Adapter repo,
instead of the adapter annotations repo.
So, no, I don't think we need to test any more than just this. :) +1 from me,
[~bdelacretaz]
> Validate the generated OSGi metatada instead of running a Sling instance
> ------------------------------------------------------------------------
>
> Key: SLING-9903
> URL: https://issues.apache.org/jira/browse/SLING-9903
> Project: Sling
> Issue Type: Improvement
> Components: Engine
> Affects Versions: Adapter Annotations 2.0.0
> Reporter: Bertrand Delacretaz
> Assignee: Bertrand Delacretaz
> Priority: Minor
>
> I took another look at the
> https://github.com/apache/sling-org-apache-sling-adapter-annotations and
> considering that the annotations act at build time only (together with the
> bnd plugin) I don't think we need to start a Sling instance to validate them.
> I'll prepare a PR that checks the XML files generated in the OSGI-INF folder,
> that should be sufficient.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
