[
https://issues.apache.org/jira/browse/MATH-291?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12751400#action_12751400
]
Luc Maisonobe edited comment on MATH-291 at 9/4/09 2:47 AM:
------------------------------------------------------------
fixed in subversion repository as of r811328
an exponential format has been used to cope with very small steps
thanks for the report and the fix proposal
was (Author: luc):
fixed in subversion repository as of r81132
an exponential format has been used to cope with very small steps
thanks for the report and the fix proposal
> ode.nonstiff.AdaptiveStepsizeIntegrator.filterStep(double h, boolean forward,
> boolean acceptSmall) can generate misleading Exception message
> --------------------------------------------------------------------------------------------------------------------------------------------
>
> Key: MATH-291
> URL: https://issues.apache.org/jira/browse/MATH-291
> Project: Commons Math
> Issue Type: Improvement
> Affects Versions: 2.0
> Reporter: Sebb
> Fix For: 2.1
>
>
> ode.nonstiff.AdaptiveStepsizeIntegrator.filterStep(double h, boolean forward,
> boolean acceptSmall) can generate misleading Exception message, for example:
> org.apache.commons.math.ode.IntegratorException: minimal step size (0)
> reached, integration needs 0
> This is because the message uses {0} and {1} to report the value of doubles.
> The code should really use a number format which allows for more decimal
> places, e.g.
> "minimal step size ({0,number,#.#########}) reached, integration needs
> {1,number,#.##########}",
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
