[postgis-devel] [PostGIS] #210: segmentation faults in lwgeom_geos.c:pgis_union_geometry_array

PostGIS trac at osgeo.org
Tue Jun 30 06:53:03 PDT 2009


#210: segmentation faults in lwgeom_geos.c:pgis_union_geometry_array
------------------------+---------------------------------------------------
  Reporter:  dfuhriman  |       Owner:  robe         
      Type:  defect     |      Status:  assigned     
  Priority:  medium     |   Milestone:  postgis 1.3.7
 Component:  postgis    |     Version:  trunk        
Resolution:             |    Keywords:               
------------------------+---------------------------------------------------
Comment (by mcayland):

 Replying to [comment:28 robe]:

 > Mark,
 >
 > Sorry I misspoke a little bit.  I think it may be fine.  I do get a
 notice but no error.  So I get a notice that says
 >
 > NOTICE:  TopologyException: found non-noded intersection between 779062
 2.92623e+06, 779069 2.92622e+06 and 779072 2.92622e+06, 778970 2.92625e+06
 779068 2.92622e+06
 > (but forgot to switch to the notice tab of PgAdmin to see)
 >
 > In the old model on 1.3 I would get a notice saying GEOS topological
 exception accompanied with a ERROR: GEOS union() threw an error!

 Hmmm. Well this bit I didn't change in the patch attached here...

 > In the PostGIS 1.4b1 I would get an error saying returned null geometry
 in addition to the topological notice.
 > In this patch I just get the topological notice -- no error which hmm in
 some cases I like better.
 >
 > So I guess I'm not sure what to consider this.  It is slightly different
 behavior, but perhaps acceptable.

 Yeah I had to change this. The old code assumed that ST_Union() returning
 NULL was an error, but of course NULL is actually valid return value when
 you perform f(NULL) on a set of NULLs...


 ATB,

 Mark.

-- 
Ticket URL: <http://trac.osgeo.org/postgis/ticket/210#comment:29>
PostGIS <http://trac.osgeo.org/postgis/>
PostGIS


More information about the postgis-devel mailing list