commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Gilles <>
Subject Re: [Numbers] Going forward
Date Wed, 18 Jan 2017 14:17:54 GMT

On Wed, 18 Jan 2017 02:46:19 +0100, Gilles wrote:
> On Tue, 17 Jan 2017 08:20:54 -0500, Raymond DeCampo wrote:
>> Let me know if there is anything I can do to help out with the 
>> transition.
> Thanks for the offer! Very much appreciated.
> I'm about to create a branch for making the project "multimodule"
> (and do some cleanup on the way).
> Sure you could help; but we are currently a bit stuck, since you do
> not have commit privilege: if I'm the only committer reviewing your
> pull requests, it will take me about the same time to do the initial
> copying myself (mainly reproducing the layout of "Commons RNG"). :-/
> So, in the meantime, I'll move the respective codes to the "core",
> "complex" and "quaternion" modules.


> Then when the layout is set up, it should be straightforward to copy
> other codes from "Commons Math" (into the respective "fraction" and
> "continuedfraction" modules).

Feel free to tackle those.
Maybe the "ContinuedFraction" class should be included in
the "fraction" module (rather then have one of its own).

> Then it will be time to decide which other functionality should move
> to the new "Commons Numbers" component.  Candidates are:
>  * o.a.c.math4.special
>  * (parts of) o.a.c.math4.analysis (e.g. the "solvers" package)

There are several in Commons Math. Shall we port only the more
robust and efficient ones (e.g. drop "RegulaFalsi")?

>  * o.a.c.math4.util.FastMath (and support classes)

In "core" or in a module of its own?

>  * o.a.c.math4.util.MathUtils
>  * (part of) o.a.c.math4.util.MathArrays

Some array utilities are definitely needed (i.e. scalar and cross
products, linear combination). [Some unit tests of "QuaternionTest"
use "Vector3D", which is out of scope.].

Is module "core" the right place for "MathArrays"?
Is the name OK?


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message