[postgis-devel] ST_Buffer in 3D supporting Z

[Paul Ramsey]

On Thu, May 3, 2012 at 4:31 PM, Sam Hariton <rotiahn at gmail.com> wrote:
> Regina,
>
> Thanks.  Unfortunately, I'm not going to make it Paris but next time your on
> the east coast I'll try to stop by (apparently I completely missed the event
> in DC last month).
>
> Is the Paris Sprint likely to result in a decision about
> GEOS/Boost/CGAL/PostGIS core/etc.?  Or is it a longer term discussion than
> that?

I don't *think* there will be any movement there, unless I seriously
misread the room.

> In the meantime, looking through the ticket list, there's a couple tickets
> (#430 and #428) that look interesting and I'd like to make a stab at.
>  What's the proper etiquette for doing so?

Check w/ Olivier quickly to make sure there's not work in progress and
then feel free. The code should live in ./liblwgeom, with SQL stubs in
./postgis. I think the extract functions are easier and would make a
better first project than something incredibly involved like IsValid
for polyhedra, which will take quite a bit of computational geometry
machinery.

p.

> Thanks
>
>  - Sam
>
> On Wed, May 2, 2012 at 4:22 AM, Paragon Corporation <lr at pcorp.us> wrote:
>>
>> Sam,
>>
>> Sam,
>> Not to my knowledge.  There are a couple of hurdles we need to jump before
>> that step.
>>
>> The big one:
>>
>> GEOS is predominantly 2-D engine:  PostGIS buffer and many other functions
>> piggy -back on that.
>>
>> so the big question is whether to expand GEOS to handle new 3d types thus
>> deviating more from its JTS roots or use another engine like Boost or CGAL.
>> The other option is just to build into PostGIS core, but not sure how much
>> effort is involved there.
>>
>> Olivier is working on much of the 3D stuff and many of the PostGIS core
>> dev are having a code sprint in Paris.
>>
>> I think there is still space left if you want to chat up with them and you
>> can make the trip.
>>
>> http://trac.osgeo.org/postgis/wiki/DevWikiEvent
>>
>> Hope that helps,
>> Regina
>> http://www.postgis.us
>>
>>
>>
>>
>> ________________________________
>> From: postgis-devel-bounces at postgis.refractions.net
>> [mailto:postgis-devel-bounces at postgis.refractions.net] On Behalf Of Sam
>> Hariton
>> Sent: Tuesday, May 01, 2012 4:06 PM
>> To: postgis-devel at postgis.refractions.net
>> Subject: [postgis-devel] ST_Buffer in 3D supporting Z
>>
>> Is anyone looking at adding the ability for st_buffer to keep Z
>> coordinates as well for 3-dimensional spaces?  If so, I'd like to help.
>>
>> The Postgis 2.0 docs state:  This function ignores the third dimension (z)
>> and will always give a 2-d buffer even when presented with a 3d-geometry.
>>
>> Right now
>>
>>> select st_astext(st_buffer(geomfromEWKT('POINT(0 0 0)'),1));
>>
>>
>> Generates a flat Polygon/Circle of radius 1 (completely dropping the z
>> dimension):
>>
>>  POLYGON((1 0,0.980785280403231 -0.195090322016128,0.923879532511287
>> -0.382683432365089,0.831469612302546 -0.555570233019602,0.707106781186548
>> -0.707106781186547,0.555570233019603 -0.831469612302545,0.382683432365091
>> -0.923879532511286,0.19509032201613 -0.98078528040323,1.61554255216634e-15
>> -1,-0.195090322016126 -0.980785280403231,-0.382683432365088
>> -0.923879532511288,-0.555570233019601 -0.831469612302546,-0.707106781186546
>> -0.707106781186549,-0.831469612302544 -0.555570233019604,-0.923879532511286
>> -0.382683432365092,-0.98078528040323 -0.195090322016131,-1
>> -3.23108510433268e-15,-0.980785280403231
>> 0.195090322016125,-0.923879532511288
>>  0.382683432365086,-0.831469612302547 0.555570233019599,-0.70710678118655
>> 0.707106781186545,-0.555570233019606 0.831469612302543,-0.382683432365094
>> 0.923879532511285,-0.195090322016132 0.98078528040323,-3.73640463187386e-15
>> 1,0.195090322016125 0.980785280403231,0.382683432365087
>> 0.923879532511288,0.5555702330196 0.831469612302547,0.707106781186545
>> 0.70710678118655,0.831469612302544 0.555570233019604,0.923879532511286
>> 0.382683432365092,0.98078528040323 0.19509032201613,1 0))
>>
>> Whereas, if it had z support it should actually generate a Polyhedral
>> Surface/Sphere of radius 1 centered on (0 0 0)
>>
>>  - Sam
>>
>>
>> _______________________________________________
>> postgis-devel mailing list
>> postgis-devel at postgis.refractions.net
>> http://postgis.refractions.net/mailman/listinfo/postgis-devel
>>
>
>
> _______________________________________________
> postgis-devel mailing list
> postgis-devel at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-devel
>