[geos-devel] Union problem
Charlie Savage
cfis at savagexi.com
Wed Jul 12 19:23:52 EDT 2006
Hi Strk,
Comments below.
> Charlie, example.cpp, *and* unit tests, all keep using the C++ API.
> SWIG is using the C API instead, so you shouldn't expect things
> to work in the exactly same way.
> I would indeed discourage you from porting unit testing and example.cpp
> but rather find new tests which might be useful for CAPI in general.
Well, maybe you can help then. What I would like is an example of two
geometries and what the expected results would be for various operations
like union, intersection, etc. That way I can verify that the results
returned through the C API are correct.
I guess I don't understand why you think example.cpp is a bad thing to
look at. I can see the geometries in their WKT format, and then I can
see what the results are when different operations happen like union, etc.
Why would the union of the same two geometries be different depending on
whether you use the C API versus the C++ API?
Anyway, I'm happy to use any examples people have. I don't think its
viable for me to make up my own examples - I don't know what the results
are supposed to look like. Thus the tests would be useless.
> The unhappy messages you're seing are in fact very happy ones :)
> It's the new code handling robustness problems by trying different
> reduction or simplification. Your logs seems to reveal that a topology
> preserving simplification with a tolerance of 0.04 did it.
> Seems nice (you should look at results to actually tell if we like it)...
Sorry, don't understand. Where is this tolerance set in doing a Union?
How do I control it through the C API?
And it then end its not happy - the results is a segmentation fault.
That of course might be a bug in the SWIG bindings.
Charlie
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3237 bytes
Desc: S/MIME Cryptographic Signature
Url : http://lists.osgeo.org/pipermail/geos-devel/attachments/20060712/21a9f429/smime.bin
More information about the geos-devel
mailing list