[postgis-users] crash with geos-2.1.3

strk at refractions.net strk at refractions.net
Wed Jan 4 05:52:04 PST 2006


GEOS version is fundamental in these report.

regress_ogc (surely misnamed) contains all function
implemented by GEOS (or JTS).

There's also a pending bug report (#14) with similar results
(worst actually, as it shows a segfault).

The bug report is with GEOS-2.1.1, which is known to be buggy.
GEOS-2.2.1 is the *latest stable* release.

Althouht MINOR version increment the ABI didn't change
between 2.1 and 2.2 (2.2 just introduced the C-API).

If GEOS-2.1.5 still gives the problem I'll be interested
in hearing bug reports. Previous versions are unsupported.

Note that building postgis against 2.2.0 will *NOT* use
the C++ API of GEOS, but rather use the new C-API interface
so problems related to C++ API use won't show up (and you
should not care actually).

--strk;

On Wed, Jan 04, 2006 at 10:16:19AM +0200, alex bodnaru wrote:
> hi all,
> 
> i am experiencing a few problems with postgis-1.1.0, on various debian
> and postgresql distributions:
> 
> with postgresql-7.4.[79], every debian distribution, see:
> /tmp/regress_regress_ogc_diff_19121
> 
> with postgresql-8.0.4, on testing, no problems encountered.
> 
> with postgresql-8.1, on testing, see:
> /tmp/regress_regress_ogc_diff_19820
> 
> willing to further test,
> 
> alex

> 35c35,36
> < polygonize_garray|GEOMETRYCOLLECTION EMPTY
> ---
> > NOTICE:  IllegalArgumentException: point array must contain 0 or >1 elements
> > polygonize_garray|

> 35c35,36
> < polygonize_garray|GEOMETRYCOLLECTION EMPTY
> ---
> > NOTICE:  IllegalArgumentException: point array must contain 0 or >1 elements
> > polygonize_garray|



More information about the postgis-users mailing list