You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Gilles Sadowski <gi...@gmail.com> on 2022/01/16 18:00:38 UTC

[Geometry] Issues (?) with the "hull" module[1]

Hello.

Could you remind me what work needs to be done so that this
functionality can make it into the next release of [Geometry]?

Thanks,
Gilles

[1] https://gitbox.apache.org/repos/asf?p=commons-geometry.git;a=tree;f=commons-geometry-hull;h=a3dd796c9c418e10768cda5998f89c19e7439f28;hb=HEAD

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org


Re: [Geometry] Issues (?) with the "hull" module[1]

Posted by Matt Juntunen <ma...@gmail.com>.
Hello.

> OK.  Did I understand correctly that the functionality is already
> there, with several implementations)?

We only have one convex hull implementation (MonotoneChain) and that
one is only for 2D. Before we release, I'd like to at least have
implementations for Euclidean 2D and 3D. I'm thinking we should design
the API so that we only present a single convex hull operation per
dimension and keep the actual implementation private, so we can
replace the algorithm if/when better alternatives become available.

Regards,
Matt

On Mon, Jan 17, 2022 at 9:10 AM Gilles Sadowski <gi...@gmail.com> wrote:
>
> Hello.
>
> Le lun. 17 janv. 2022 à 14:16, Matt Juntunen
> <ma...@gmail.com> a écrit :
> >
> > Hi Gilles,
> >
> > The main things I want to do on the hull module are
> > - Review the public API to see if any changes need to be made. (ex:
> > make AklToussaintHeuristic internal)
>
> Good suggestion; I'll minimize the public API.[1]
>
> > - Add QuickHull implementations for 2D/3D (GEOMETRY-110). This is
> > currently blocked due to the need for a PointMap implementation, which
> > I am working on now in GEOMETRY-142.
>
> OK.  Did I understand correctly that the functionality is already
> there, with several implementations)?  If so, we could in effect
> make all those private, and select a different alternative whenever
> a more performant algorithm is available.
>
> Best,
> Gilles
>
> [1] I've created https://issues.apache.org/jira/browse/GEOMETRY-144
>
> > On Sun, Jan 16, 2022 at 12:59 PM Gilles Sadowski <gi...@gmail.com> wrote:
> > >
> > > Hello.
> > >
> > > Could you remind me what work needs to be done so that this
> > > functionality can make it into the next release of [Geometry]?
> > >
> > > Thanks,
> > > Gilles
> > >
> > > [1] https://gitbox.apache.org/repos/asf?p=commons-geometry.git;a=tree;f=commons-geometry-hull;h=a3dd796c9c418e10768cda5998f89c19e7439f28;hb=HEAD
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> For additional commands, e-mail: dev-help@commons.apache.org
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org


Re: [Geometry] Issues (?) with the "hull" module[1]

Posted by Gilles Sadowski <gi...@gmail.com>.
Hello.

Le lun. 17 janv. 2022 à 14:16, Matt Juntunen
<ma...@gmail.com> a écrit :
>
> Hi Gilles,
>
> The main things I want to do on the hull module are
> - Review the public API to see if any changes need to be made. (ex:
> make AklToussaintHeuristic internal)

Good suggestion; I'll minimize the public API.[1]

> - Add QuickHull implementations for 2D/3D (GEOMETRY-110). This is
> currently blocked due to the need for a PointMap implementation, which
> I am working on now in GEOMETRY-142.

OK.  Did I understand correctly that the functionality is already
there, with several implementations)?  If so, we could in effect
make all those private, and select a different alternative whenever
a more performant algorithm is available.

Best,
Gilles

[1] I've created https://issues.apache.org/jira/browse/GEOMETRY-144

> On Sun, Jan 16, 2022 at 12:59 PM Gilles Sadowski <gi...@gmail.com> wrote:
> >
> > Hello.
> >
> > Could you remind me what work needs to be done so that this
> > functionality can make it into the next release of [Geometry]?
> >
> > Thanks,
> > Gilles
> >
> > [1] https://gitbox.apache.org/repos/asf?p=commons-geometry.git;a=tree;f=commons-geometry-hull;h=a3dd796c9c418e10768cda5998f89c19e7439f28;hb=HEAD

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org


Re: [Geometry] Issues (?) with the "hull" module[1]

Posted by Matt Juntunen <ma...@gmail.com>.
Hi Gilles,

The main things I want to do on the hull module are
- Review the public API to see if any changes need to be made. (ex:
make AklToussaintHeuristic internal)
- Add QuickHull implementations for 2D/3D (GEOMETRY-110). This is
currently blocked due to the need for a PointMap implementation, which
I am working on now in GEOMETRY-142.

Regards,
Matt J

On Sun, Jan 16, 2022 at 12:59 PM Gilles Sadowski <gi...@gmail.com> wrote:
>
> Hello.
>
> Could you remind me what work needs to be done so that this
> functionality can make it into the next release of [Geometry]?
>
> Thanks,
> Gilles
>
> [1] https://gitbox.apache.org/repos/asf?p=commons-geometry.git;a=tree;f=commons-geometry-hull;h=a3dd796c9c418e10768cda5998f89c19e7439f28;hb=HEAD
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
> For additional commands, e-mail: dev-help@commons.apache.org
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@commons.apache.org
For additional commands, e-mail: dev-help@commons.apache.org