On Tue, Jan 06, 2009 at 10:26:34PM +0300, Ivan Illarionov wrote: > On Tue, Jan 6, 2009 at 9:48 PM, Mark de Wever <[email protected]> wrote: > > > Sorry but this doesn't answer any of my questions. And I don't see what > > Linus's dislike with C++ has to do with Wesnoth. Some of his reasons to > > dislike C++ also apply to Python. After reading Linus's mail I see the > > solution would be to switch to C instead of Python ;-) > > The reasons apply to the questions why I use plain C/Python API and > dislike boost or SWIG or anything else. But instead of using boost or SWIG you think it would be good to modify the Wesnoth code to make it easier to work with the Python/C API. I _really dislike_ the idea of making a lot of modifications to the new GUI code.
> The level of information hiding afforded by this tools also hides > potential problems. Yes, they will "help" me program, but don't > give me the level of flexibility that I need at this stage. Information hiding is often called abstraction. Please give some details about in what way boost or SWIG limits your flexibility. From what I heard thusfar the Python/C API is not flexible enough; you want to modify the new GUI code to make it easier to use with the Python/C API. > Plain Python/C API is also faster (I consider speed important for the > core game engine as opposed to gameplay layer) and more Python > friendly because Python itself is written in C. If you claim that Python/C API is faster as boost or SWIG, please post some benchmarks to backup those claims. I don't buy those claims without a benchmark. -- Regards, Mark de Wever aka Mordante/SkeletonCrew _______________________________________________ Wesnoth-dev mailing list [email protected] https://mail.gna.org/listinfo/wesnoth-dev
