[
https://issues.apache.org/jira/browse/MATH-894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13497655#comment-13497655
]
Gilles commented on MATH-894:
-----------------------------
It seems rather far-fetched to imagine that an application would rely on the
fact that "clear()" resets the size of the array; this is/was a quite deeply
internal design choice.
IIRC, some slightly incompatible functional changes are allowed if they improve
the intended usage. Let's say that if someone complains, I'll prepare a 3.1.1
release :).
As for "double" vs "float", it's an oddness since there is no reason to not use
"double" for those two fields. I don't think that it will entail any mess; I'll
add new constructors (with "double" args) and deprecate the old ones. Unless
I'm missing something, users will have nothing to do.
> Cleanup of "ResizableDoubleArray" ("o.a.c.m.util")
> --------------------------------------------------
>
> Key: MATH-894
> URL: https://issues.apache.org/jira/browse/MATH-894
> Project: Commons Math
> Issue Type: Improvement
> Affects Versions: 3.0
> Reporter: Gilles
> Assignee: Gilles
> Priority: Minor
> Fix For: 3.1, 4.0
>
> Attachments: MATH-894.patch
>
>
> Method "addElement" should not call "contract()", as it will almost always
> make the storage array shrink on the first call since it will be considered
> "too big" (for just storing one element).
> See discussion about this change in MATH-757.
> Miscellaneous code modifications are also proposed in order to improve
> encapsulation.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
