[
https://issues.apache.org/jira/browse/MYFACES-2537?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12829950#action_12829950
]
Leonardo Uribe commented on MYFACES-2537:
-----------------------------------------
Thinking more about it, I like the old algorithm version, because it handles
better "before others" and "after others" rules. Since it try to resolve nodes
from lower to upper complexity, the "others" group are handled better. The
alternative could find cases like presented before, so I would like myfaces try
first with the previous version and if a cyclic reference is found, try the
topological sorting or other alternative to be proposed.
> FacesConfigurator.sortRelativeOrderingList() algorithm is broken trying to
> resolve some examples
> ------------------------------------------------------------------------------------------------
>
> Key: MYFACES-2537
> URL: https://issues.apache.org/jira/browse/MYFACES-2537
> Project: MyFaces Core
> Issue Type: Bug
> Components: JSR-314
> Affects Versions: 2.0.0-beta
> Reporter: Leonardo Uribe
> Assignee: Leonardo Uribe
>
> Curtiss Howard put this comment on dev list
> FacesConfigurator.sortRelativeOrderingList() is failing on a rather
> simple case:
> A after B
> B before C
> C before A
> The expected faces-config ordering is B-C-A, but instead
> sortRelativeOrderingList() is detecting a circularity.
> The algorithm proposed fails because it is not able to process the nodes in
> the correct order (the algorithm assign a weight equal to all nodes, so it
> fails when try to order them in a psedo "postorder" form).
> It is faster and better try another algorithm. The current one works in all
> tests done at this moment but this case makes it fail without any possible
> workaround.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
