You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@ant.apache.org by "Michael Fraenkel (JIRA)" <ji...@apache.org> on 2009/04/11 15:35:20 UTC

[jira] Closed: (IVY-1062) ChainResolver broke with dynamic revisions

     [ https://issues.apache.org/jira/browse/IVY-1062?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Michael Fraenkel closed IVY-1062.
---------------------------------

    Resolution: Fixed

Never mind.  We implemented a resolver which didn't look at the previous resolved revision.

> ChainResolver broke with dynamic revisions
> ------------------------------------------
>
>                 Key: IVY-1062
>                 URL: https://issues.apache.org/jira/browse/IVY-1062
>             Project: Ivy
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 2.0
>            Reporter: Michael Fraenkel
>
> We have defined a ChainedResolver with 3 FileSystemResolvers.
> A resolve to locate a resource foo with revision [1.0.0.0, ) is invoked.
> ChainResolver.getDependency() is invoked and the first resolver is called.  (line 126)
> A ResolvedModuleRevision is returned.
> We have returnFirst set on the ChainedResolver which really won't affect anything yet.
> The second resolver is called and nothing is returned.
> The third resolver is invoked.
> Now line 125, sets the current resolved module revision to null wiping out the one that was returned.
> The third resolver finds nothing.
> We have no more resolvers.
> We exit the resolver loop, knowing nothing.
> mr == null, but we have no errors which is a good thing.  (line 141)
> resolved == mr since they are both null at this point (line 162)
> The ChainedResolver returns null as the ResolvedModuleRevision which is obviously wrong.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.