[fdo-dev] Classes and properties in SHP providers

[Frank Warmerdam]

Dan Stoica wrote:
> I believe the number of capabilities will just keep growing. Rationale:
> it is absurd to think otherwise :-).
> 
> Now, I believe there was a better method to support capabilities:
> instead of "SuportsThat()" methods we could have just one method
> "Supports( that )" on the connection, where "that" is an enum. Some
> advantages:
> - way less methods on the interfaces
> - makes the maintance less painful (right now we need to touch ALL
> providers, just to say "not supported").

Amen Brother!

I agree that having to extend the API/ABI for every capabilities test
is overkill.  OGR has a TestCapability() method on core classes
with a #defined name for each capability.  I an enumerated list of
capabilities and a single method on each of the various Capabilities
class would help quite a bit.

Well, at least for capabilities that reduce to a simple yes/no answer.
For stuff like returning maximum data width, or testing a particular
raster configuration a more specific method is needed.

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