[
https://issues.apache.org/jira/browse/SLING-4734?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14725444#comment-14725444
]
Radu Cotescu commented on SLING-4734:
-------------------------------------
We should also optimise the {{ModelFactoryUseProvider}} to fail as early as
possible, therefore I'd call this line \[0\] immediately the resource and class
object can be obtained.
\[0\ -
https://github.com/apache/sling/blob/90d2ed9e42deb144a7f6e1610871e72726cd810a/bundles/scripting/sightly/models-use-provider/src/main/java/org/apache/sling/scripting/sightly/models/impl/ModelFactoryUseProvider.java#L98
> Sightly should try to instantiate Sling Models first via
> SlingHttpServletRequest then via Resource and not the other way around
> -------------------------------------------------------------------------------------------------------------------------------
>
> Key: SLING-4734
> URL: https://issues.apache.org/jira/browse/SLING-4734
> Project: Sling
> Issue Type: Improvement
> Components: Scripting
> Affects Versions: Scripting Sightly Models Use Provider 1.0.0
> Reporter: Konrad Windszus
> Fix For: Scripting Sightly Models Use Provider 1.0.0
>
>
> Currently the {{JavaUseProvider}} first tries to instantiate Sling Models via
> the {{Resource}} then via the {{SlingHttpServletRequest}}. I would argue that
> the inverse order makes more sense, because more injections are supported via
> the {{SlingHttpServletRequest}} for Sling Models.
> This would only matter though, where the same class would be adaptable from
> both {{Resource}} and {{SlingHttpServletRequest}}. In that case though the
> instantiation via the {{SlingHttpServletRequest}} would support more
> injections (e.g. the SlingObjectInjector supports much more with the
> {{SlingHttpServletRequest}}, compare with SLING-4083)
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
