[shrinkwrap-issues] [JBoss JIRA] (SHRINKDESC-163) Deal with descriptors with multiple possible roots better

Ralf Battenfeld (JIRA) issues at jboss.org
Sun Oct 11 10:05:00 EDT 2015


    [ https://issues.jboss.org/browse/SHRINKDESC-163?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13117224#comment-13117224 ] 

Ralf Battenfeld commented on SHRINKDESC-163:
--------------------------------------------

I worked a little bit over the weekend. What I see is the following:

* introducing a marker interface feature as configurable entry in the maven-metadata-plugin. Solves Bob's minimal wish. We do have something that allows to put common things together but the issue is that the top level element is already different.
* Introducing a wrapper descriptor wrapping descriptors internally. You can then at least load a file and you get then the right descriptor back. We may can then extend this wrapper with common stuff for all internal descriptors.

Makes this sense for you?

> Deal with descriptors with multiple possible roots better
> ---------------------------------------------------------
>
>                 Key: SHRINKDESC-163
>                 URL: https://issues.jboss.org/browse/SHRINKDESC-163
>             Project: ShrinkWrap Descriptors
>          Issue Type: Feature Request
>    Affects Versions: 2.0.0-alpha-8
>            Reporter: Toby Crawley
>
> When implementing SHRINKDESC-162, I created three different
> descriptors for module.xml ({{ModuleDescriptor}},
> {{ModuleAliasDescriptor}}, {{ModuleAbsentDescriptor}}) corresponding to
> the three root elements listed in the xsd ({{module}}, {{module-alias}},
> {{module-absent}}). This works fine if you are generating a module.xml,
> but if you are loading in an existing file, you don't know without
> looking at the file contents to know which type to instantiate.
> It would be swell if there was a way for descriptors to support
> multiple roots and do the correct thing. Or does that already exist
> and I missed it?



--
This message was sent by Atlassian JIRA
(v6.4.11#64026)


More information about the shrinkwrap-issues mailing list