[Liblas-devel] .net/mono bindings in c#
Mateusz Loskot
mateusz at loskot.net
Wed Aug 6 10:23:09 EDT 2008
Martin Rodriguez wrote:
> Hi all:
>
> I am working in .net/mono bindings to liblas and i have something like
> 80% work done,
Martin,
Great news!
> but i have several problems:
>
> - I have created a successful mingw compilation without use
> CAPI module. This module use a macro _MSC_VER in the las_config.h header
>
> /* FIXME - Ticket #23 */
>
> #ifdef _MSC_VER
>
> #define PACKAGE_VERSION "1.0.0b1"
>
> #endif
>
> - CAPI use this macro only to return in a function. I believe
> in liblas purposes we could do something like this:
>
> - defined(__CYGWIN__) || defined(__MINGW32__) ||
> defined(_MSC_VER)
Good point. Could you please report it as a Ticket [1] and attach
proposed changes in all places you need?
[1] http://liblas.org/
> - C++ API compile and works fine, but I am worried because I see the
> macro in some other headers like bin endian/little endian module.
You mean the _MSC_VER macro used in there?
If yes, so while reporting a ticket, pelase specify all places where you
propose to change it.
> - Anyway I am using your DLL for .NET, but it would be great
> for me work with my DLL. Besides Mingw binaries are very useful for me
> because almost all APIs I use in C/C++ are ready to mingw (gtk,gtkmm,
> inscape, gimp, vala, ruby, etc, etc) . There many only 2 libraries I use
> for VS like Mysql and postgree, but mingw have a DLL converter tool to
> import them. .NET have the advantage to find entry point in mingw and vc
> DLL in the same way.
Understood.
> - I am working in Setters and getters in c# and I see some rare
> things in the CAPI module.
>
> LAS_DLL uint8_t LASHeader_GetDataFormatId(const LASHeaderH hHeader);
>
> - LAS_DLL LASError LASHeader_SetDataFormatId(const LASHeaderH
> hHeader, int value);
>
> - Why do you have a kind of data to c++ to the getter and other
> to the setter???. In c++ this is not a problem because the getters and
> setter are visible, but in .NET not (only in compiled code). I saw like
> 3 functions like that.
You mean the uint8_t vs int ?
Yes, it's incorrect and I believe it should be changed to uint8_t.
Howard, do you have any objections about this fix?
Martin, if I may ask, are you going to publish your C# bindings as Open
Source?
Best regards,
--
Mateusz Loskot, http://mateusz.loskot.net
Charter Member of OSGeo, http://osgeo.org
More information about the Liblas-devel
mailing list