wxDC reorganisation

[Robert Roebling]

 Hi all,

I have been thinking about how to add support
for Cairo drawing to wxGTK's wxDC code. One
way would be a pimpl mechanism for wxGTK alone
without touching any other code at all. It is
quite obvious that a more general appraoch
would be better, basically making it possible
to plug-in drawing implementation into a wxDC
interface. 
In the end, there will only be one wxDC class
(in base) and the wxDCBase class will dissappear
(or rather be merged with that). Then some
code can e.g. test for the presence of Cairo
(on any platform) and use that by using the
Cairo backend. Does anyone think this should
not be done or could be better done? In the
transition phase, there is likely going to be
some breakage, but it will also unify some code
and hopefully remove some redundant code. I just
wanted to clean up the SVG wxDC code and realized
how much clean up is needed and that the current
wxDC "architecture" invites chaos

  Robert