[ https://issues.apache.org/jira/browse/MATH-758?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13221445#comment-13221445 ]
Sebb commented on MATH-758: --------------------------- LaguerreSolver.complexSolver - could this be private and final? Dfp fields exp/mant/nans/sign - looks like these could be private with getters, apart from nans which is written by subclasses. AbstractFormat.setDenominatorFormat() AbstractFormat.setNumeratorFormat() - are these needed (only used in test code)? If not, the underlying fields could be made final NordsieckStepInterpolator.stateVariation is only used internally AFAICT, so could be private FirstMoment.n has a getter, and is not updated externally, so could be private FourthMoment.m4 is only read externally by Kurtosis, so could be private (there is a getter) The fields m1, m2, m3 are only ever written by the xxxMoment classes, which set them to zero. Each class could have a protected zero() method, which could be chained in the same way as the clear() method. Other classes could then use the getters and m1-m4 could be private. [To be continued] > Fields which could be private and/or final > ------------------------------------------ > > Key: MATH-758 > URL: https://issues.apache.org/jira/browse/MATH-758 > Project: Commons Math > Issue Type: Bug > Reporter: Sebb > > BaseAbstractUnivariateIntegrator has several fields that are not currently > changed after construction and could be final: > protected double absoluteAccuracy; > protected double relativeAccuracy; > protected int minimalIterationCount; > protected Incrementor iterations; > protected Incrementor evaluations; > These all have getters as well, so could also be made private. -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira