[wx-discuss] wxWidgets version 3?

[Vadim Zeitlin]

On Wed, 15 Feb 2006 15:45:31 -0800 Robin Dunn <robin at alldunn.com> wrote:

RD> Goal #1:  wxTNG shall make it very easy to create language bindings such 
RD> as wxPython.

 It's not very surprizing that you should say this, of course, but it can
potentially be quite a big problem because some of the other changes risk
to go in the opposite direction.

RD> a. Reduce the reliance on virtual methods to be as small as possible. 

 In one possible design I'm thinking of there are no virtual functions at
all so this might make you happy.

RD> b. "Clean" headers without a lot of heavy macro use or complex C++ 
RD> cruft.

 OTOH just about anything I think of for wxTNG involves using complex C++
which will probably be unparsable by SWIG (unless its template support has
improved beyong my wildest dreams). I really don't know how to reconcile
wxTNG plans with SWIG features. I'm very curious how does Stefan deal with
this.

 Personally I think that maybe the only way to make it work is to have a
wx2-like low-level library with wxTNG API on top of it. This would allow to
continue using SWIG (on low-level library headers) and also keep some
(most?) compatibility with wx2 API, too.

RD> c. Most C++ class instances should have a user-data pointer that is not 
RD> used for anything in the framework.

 This is a very C-ish approach, in C++ there are really better ways. But I
think that this is not very important, in the worst case you can just use a
map to associate user-data with any pointer anyhow.

RD> d. It would be nice if there was a separation of __WXDEBUG__ and 
RD> wxASSERT

 This is pretty minor as well but it's probably still worth it to add to
the page Ryan mentioned:

	http://www.wxwidgets.org/wiki/index.php/WxWidgets3

 Regards,
VZ


---------------------------------------------------------------------
To unsubscribe, e-mail: wx-discuss-unsubscribe at lists.wxwidgets.org
For additional commands, e-mail: wx-discuss-help at lists.wxwidgets.org