[gdal-dev] Should we use boolean type at the SWIG interface?

[Tamas Szekeres]

Frank,

I'd rather not to change the API as I've suggested before because many
existing code may rely on it. By using this change the existing
scripts should use the !=0 suffix in the conditions and this change
may be confusing.

What I would propose now is to change the related code according to
the following example:

bool TestCapability (const char *cap) {
    return OGR_Dr_TestCapability(self, cap) > 0? true: false;
  }

How about this way?

Tamas



2008/7/13 Frank Warmerdam <warmerdam at pobox.com>:
> Tamas Szekeres wrote:
>>
>> Devs,
>>
>> By using boolean type in the SWIG bindings I currently get a number if
>> warnings when compiling the generated code:
>>
>> http://buildbot.osgeo.org:8500/builders/szekerest-vc71-full/builds/36/steps/csharp-make/logs/stdio
>>
>> Since there are no boolean value exposed by the gdal dll I suppose we
>> can use integers instead of booleans at the wrappers as well.
>>
>> Any objection to relpace 'bool' with 'int' in ogr.i?
>
> Tamas,
>
> I believe it would be simplest to expose boolean results as a int.
>
> I see that the C API already treats the return result of TestCapability()
> as an int.  I'm not sure why it was done differently in the swig include
> file.
>
> 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
>
>