[Gdal-dev] GDALDataset as a Base Class

[Frank Warmerdam]

Ivan Lucena wrote:
> Matt, All,
> 
> You are not alone, I don't mind use the pointer either. See in the
> attachment an initial attempt to implement Michael Goodchild's Spatial
> Model (Well, at least GetExtent() is working identically for
> Fields[GDAL] and Objects[OGR] :-).
> 
> But what bugs me a little bit is the proliferation of wrapper and
> facades for GDAL that *we* are creating. For example, I am following the
> idea of VB6 GDAL wrapper and implementing a Delphi wrapper!
> 
> And I also believe that we can count the wrappers automatic created by
> swig because they also produce slightly different interfaces that
> requires especial documentation (and a separated support group from the
> gdal-dev list).

Ivan,

It is certainly impractical to support a huge number of different bindings
for GDAL in different languages well.  With the swig languages, my hope
was that a objects would appear similarly in all languages so it would be
easier to "document once", or even to relate things back to the core C++ API.

But I don't have a solution, and I'm not going to prevent people from doing
bindings just because they are hard to document.  But currently I think very
few of the bindings can be considered well supported.

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