[postgis-users] Feautures omitted from ST_Buffer operation

Martin Davis mbdavis at refractions.net
Thu Mar 26 08:57:10 PDT 2009

I don't think this can be the problem - a buffer of a linestring should 
still produce a polygon.

This is probably a bug in the GEOS buffer algorithm.  It would help if 
you posted the WKB or WKB for the geometry.

Chris Hermansen wrote:
> This sounds to me like you might be buffering a linestring and not a
> polygon.  Is that possible?  Ie are the countries represented as
> outlines (linestrings) rather than polygons?
> Roger André wrote:
>> Hi All,
>> I'm having some problems with features that are being returned from an
>> ST_Buffer operation in PostGIS.  When I try to buffer the extents of
>> Papua New Guinea, a large region in the middle of its extents is
>> omitted.  You can see a visual representation of this at
>> http://lagerratrobe.blogspot.com/2009/03/problem-with-postgis-stbuffer.html
>> If I just select the country itself, I get back all of the features
>> that I'm expecting.  In the image I've posted above, the black area
>> indicates the results of that query.
>> I am using postgresql-8.3.5, and postgis-1.3.3.  My select statements
>> is as follows:
>>     select name, Buffer(the_geom, .12) from coastlines where
>> iso_3_code = 'PNG';
>> I've tried this using both a free countries shapefile, as well as an
>> ESRI one, and get the same results.  I'm a bit baffled, and unsure if
>> I have a problem with both data sets, or in my query, or...?  If
>> someone else has had a similar problem in the past, and can shed some
>> light on it, I'd greatly appreciate it.
>> Thanks in advance,
>> Roger
>> --
>> ------------------------------------------------------------------------
>> _______________________________________________
>> postgis-users mailing list
>> postgis-users at postgis.refractions.net
>> http://postgis.refractions.net/mailman/listinfo/postgis-users

Martin Davis
Senior Technical Architect
Refractions Research, Inc.
(250) 383-3022

More information about the postgis-users mailing list