[
https://issues.apache.org/jira/browse/MATH-441?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12931825#action_12931825
]
Gilles commented on MATH-441:
-----------------------------
I've removed most occurrences from package {{optimization}} in revision 1034996.
However, there is a problem with the "MultiStart..." classes: e.g. in
{{BaseMultiStartMultivariateVectorialOptimizer}}, line 150, the code catches
{{FunctionEvaluationException}}. This contradicts the promise to users that the
new {{MathUserException}} won't be caught in the CM code.
Moreover, IMO, the semantics of catching a user-generated exception is too-high
level here (i.e. we assume that the exception is "normal" for the user of the
multi-start optimizer, whereas it could well be a programming error that will
be swallowed by the CM code...). What I mean is that we shouldn't second-guess
the user on whether problems in function evaluation are expected or not:
multi-start should not be construed as allowing invalid starts.
In that case, the fix is simply to remove the catch block.
What do you think?
> Remove occurences of "FunctionEvaluationException"
> --------------------------------------------------
>
> Key: MATH-441
> URL: https://issues.apache.org/jira/browse/MATH-441
> Project: Commons Math
> Issue Type: Task
> Reporter: Gilles
> Assignee: Gilles
> Priority: Minor
> Fix For: 3.0
>
>
> As per the issue MATH-440, {{FunctionEvaluationException}} is now redundant
> and is going to removed.
> We should remove the "throws" clauses that refer to it from all the methods'
> signature.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
