[postgis-devel] Segfault fixes in 1.0 branch [was: Trouble with AsText(ExteriorRing())]

[strk at refractions.net]

I found the bug, both exteriorring() and segmentize() are
affected. The bug has been introduced in the process of
fixing memory alignment problems. This calls for 1.0.6
but this time please help testing.

To test candidate 1.0.6 you should fetch the code tagged
as 'pgis_1_0':

export CVSROOT=:pserver:strk at postgis.refractions.net:/home/cvs/postgis
cvs co -r pgis_1_0 postgis

TIA

--strk;

On Sat, Nov 26, 2005 at 02:44:53PM +0100, Diez B. Roggisch wrote:
> strk at refractions.net wrote:
> >No problems with regress test.
> >Can you provide a testcase for me to check at ?
> >I would need the smallest possible dataset,
> >in this case the Geometry which triggers the
> >segfault when feed to the IsSimple() function.
> >
> >To find out which one it is I suggest you
> >define a cursor for select id,IsSimple(the_geom)
> >and fetch. Last seen id+1 will be the faulting
> >geometry.
> >
> >Once isolated create a table as select id,the_geom
> >where id = <found_id> and produce a shapefile from
> >it using pgsql2shp. Send the shapefile to me and
> >I'll see if I can reproduce it.
> 
> Here it comes. Please notice that after upgrading libgeos the probkme is 
> _not_ IsSimple anymore, but ExteriorRing (as my changed subject said). 
> However, the symptoms are all the same.
> 
> Thanks for your support - it is highly appreciated!
> 
> Regards,
> 
> Diez


> _______________________________________________
> postgis-users mailing list
> postgis-users at postgis.refractions.net
> http://postgis.refractions.net/mailman/listinfo/postgis-users


-- 

 /"\    ASCII Ribbon Campaign
 \ /    Respect for open standards
  X     No HTML/RTF in email
 / \    No M$ Word docs in email