[
https://issues.apache.org/jira/browse/MATH-719?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13163737#comment-13163737
]
Gilles commented on MATH-719:
-----------------------------
bq. Sorry if this doesn't belong here.
Indeed, you'd better bring this kind of issue to the "dev" ML. :)
The more so that there have been recent discussions about changing the matrix
API and decisions ought to be made quite soon now.
> Strange deprecations in API
> ---------------------------
>
> Key: MATH-719
> URL: https://issues.apache.org/jira/browse/MATH-719
> Project: Commons Math
> Issue Type: Bug
> Affects Versions: 2.0, 2.1, 2.2
> Reporter: Peter Bloem
> Priority: Minor
> Labels: api-change, deprecated
>
> Sorry if this doesn't belong here. I couldn't find any sort of mailing list
> or other feedback mechanism on the website.
> RealMatrix has some very odd deprecations. In particular inverse(),
> getDeterminant() and isSingular(). The last has the message:
> bq. Deprecated. as of release 2.0, replaced by the boolean negation of new
> LUDecompositionImpl(m).getSolver().isNonSingular()
> That's an implementation, not an interface. The whole point of having an
> interface is that
> * I can query whether a matrix is singular withou having to know about
> LUDecompositions
> * You guys can change the implementation of isSingular() if something better
> pops up without us guys having to change our code.
> I'm not using these methods now, because they're deprecated, but I've
> basically recreated them in as static methods in a utility class. Wouldn't it
> be much better to just put code from the deprecation message into the method
> and remove the deprecation?
--
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
