[postgis-users] ST_UNION drops M coordinate

Paragon Corporation lr at pcorp.us
Wed Aug 5 16:08:36 PDT 2009


The closest thing we have to a table is in the special index of the help.  I
think it covers all the functions that support this though we were focusing
more on Z coordiante than M.  So probably M we missed and should be more
specific about.

http://postgis.refractions.net/documentation/manual-svn/ch08.html#PostGIS_3D
_Functions

I like the idea of a chart though.  Have to think about a way to automate
that.  

Thanks,
Regina



-----Original Message-----
From: postgis-users-bounces at postgis.refractions.net
[mailto:postgis-users-bounces at postgis.refractions.net] On Behalf Of Paul
Ramsey
Sent: Wednesday, August 05, 2009 1:52 PM
To: PostGIS Users Discussion
Subject: Re: [postgis-users] ST_UNION drops M coordinate

No, sorry no table. In general:

"interesting" things (containment tests, calculations of new derived
shapes) only take X and Y into account.
constructive functions (buffer, union, difference) that make new geometries
pass Z through and for new points try to interpolate a "reasonable" Z value
from the inputs constructive functions do not pass M through

So the most you can expect from higher dimension handling is that they are
not stripped and are hopefully correctly interpolated on introduced
vertices.

P.

On Wed, Aug 5, 2009 at 10:46 AM, Satish Murthy<satish_m at hotmail.com> wrote:
> Thanks Paul.
>
> I also noticed that some other functions too have this limitation 
> (such as ST_CONTAINS). Is there a table/matrix somewhere that lists 
> which function supports XYZM?
>
> Satish
>
>> Date: Wed, 5 Aug 2009 10:36:46 -0700
>> From: pramsey at opengeo.org
>> To: postgis-users at postgis.refractions.net
>> Subject: Re: [postgis-users] ST_UNION drops M coordinate
>>
>> Yes, we just don't support the higher dimension in the operation. 
>> GEOS was extended to 3 dimensional handling some time ago, but never 
>> got pushed farther when we added M at postgis 1.0. Probably fixable, 
>> but with a reasonable amount of careful work to catch every case.
>>
>> P
>>
>> On Wed, Aug 5, 2009 at 10:22 AM, Satish Murthy<satish_m at hotmail.com>
>> wrote:
>> >
>> > When I do this (notice the second ST_COLLECT):
>> > select ST_AsEWKT (ST_COLLECT (geom)) from ( select 
>> > (ST_DUMP(geom)).geom from (select p_geom as geom from ( select 
>> > ST_COLLECT (geom_col, st_GeomFromEWKT('Point(1 -1 0 10)')) as 
>> > p_geom from tab) a )as b ) as c I get:
>> > "MULTIPOINT(71.1 -42.31 10 10,1 -1 0 10,1 -1 0 10)"
>> > This is fine, except that "1 -1 0 10" gets repeated.
>> >
>> > BUT---
>> > When I do (I use ST_UNION here):
>> > select ST_AsEWKT (ST_COLLECT (geom)) from ( select 
>> > (ST_DUMP(geom)).geom from (select p_geom as geom from ( select 
>> > ST_UNION (geom_col, st_GeomFromEWKT('Point(1 -1 0 10)')) as p_geom 
>> > from tab) a )as b ) as c I get this:
>> > "MULTIPOINT(1 -1 0,71.1 -42.31 10)"
>> > No repetition, but the M coordinate is missing.
>> >
>> > Any pointers on what's going wrong here?
>> >
>> >
>> > ________________________________
>> > Are you an untamed, bizarre or daring explorer? Find out now! Drag n'
>> > drop
>> > _______________________________________________
>> > postgis-users mailing list
>> > postgis-users at postgis.refractions.net
>> > http://postgis.refractions.net/mailman/listinfo/postgis-users
>> >
>> >
>> _______________________________________________
>> postgis-users mailing list
>> postgis-users at postgis.refractions.net
>> http://postgis.refractions.net/mailman/listinfo/postgis-users
>
> ________________________________
> Date India's Teen Diva. Find out if you hit it off with her or are her 
> fiery opposite. Drag n' drop 
> _______________________________________________
> postgis-users mailing list
> postgis-users at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-users
>
>
_______________________________________________
postgis-users mailing list
postgis-users at postgis.refractions.net
http://postgis.refractions.net/mailman/listinfo/postgis-users





More information about the postgis-users mailing list