[Gdal-dev] Re: C# Geometry Classes?

[Frank Warmerdam]

Tamas Szekeres wrote:
> Hi Folks,
> 
> It would be a significant effort to discover which functionality
> cannot be treated this way. For example it doesn't make too much sense
> to support IsPointOnSurface on a Geometry class.
> Sooner or later we should harmonize the C++ and the SWIG API or
> support a separate documentation for the SWIG interface. Otherwise the
> user will always search for specific methods described at the C++
> documentation but missing from the SWIG interface.

Tamas,

I would agree that at the least we need to document the C API in an
accessable manner.  Mateusz, please add this to your TODO list, and chat
with me when you are ready to start so we can ensure it is findable.

Believe it or not, for many purposes I prefer the OGR C API for geometry
due to the more generalized model for accessing points and children.  For
the GMT vector format driver I have recently committed in trunk I use the
C API internally for decomposing geometries for writing to the file since
I don't have to fool around with a lot of casting to specific types.

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | President OSGeo, http://osgeo.org