[postgis-users] ST_Insersection problem

[G. van Es]

Sandro, thanks for your reply. You're right about "folding postgis 7 times" :-)

I'm having difficulties in isolating the offending geometries, which is mainly due to my lack of knowledge.

When I look at the location at 59647.5 392932 there is no intersection. Only when I use buffer(geom,1) there are two objects, one of them being a funny line of three points but still identified by PG as polygon.  


So I'm a bit lost now. I will continue to try to isolate a few features to consistently reproduce the problem.

Ge



________________________________
From: Sandro Santilli <strk at keybit.net>
To: G. van Es <gves2000 at yahoo.com>; PostGIS Users Discussion <postgis-users at postgis.refractions.net>
Sent: Wednesday, September 21, 2011 8:52 AM
Subject: Re: [postgis-users] ST_Insersection problem

On Tue, Sep 20, 2011 at 11:01:23PM -0700, G. van Es wrote:
> Well, ST_SnapToGrid seems to work a little better we get now TopologyException: side location conflict at 59647.5 392932
> 
> Still GOES seems to crash on Intersection()...... 
> It's a bit funny since all data is produced by PostGis and now PostGis
> can't handle it's own data.

Well, it's a robustness problem.
Try folding a piece of paper for more than 7 times. It's a similar issue.

If you can isolate the offending operands I might take a closer look.
Please file a ticket for it.

--strk;

  ()   Free GIS & Flash consultant/developer
  /\  http://strk.keybit.net/services.html
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/postgis-users/attachments/20110921/18551d00/attachment.html>