I'm not sure I entirely agree with Jeff's sentiments, although it would make things painful.> Since Evo is a GNOME app, not a GTK app, you'd need, I think, > all of the GNOME libs ported to MSWin. Yes. porting the GNOME libs to windows is much work and in some cases might be not easy, probably. But would it be possible to remove GNOME dependencies and substitute the components in question by things that are available on Windows, either through the Windows API or other porting projects? Or are there to many dependencies on GNOME and it would not be possible to do the evolution port without the GNOME libs port?
It would be a major undertaking. Particularly if you wanted a more native feel to it, it would essentially be a total rewrite. If you just ran it under an X server it would just be a lot of work.
Most of the nasty bulky code is either:
- very posix i/o specific, filesystem semantics specific too.
- posix thread specific
- fairly X specific (selections, drag and drop, i don't know how well gtk+ maps to windows maybe it makes it easy)
Although its been so long since I did anything on windows I have no idea what services are available now.
Miguel has the wild idea that we're all going to throw all the code away and rewrite it in c# anyway, providing a cross-platform port 'for free'. We could have done that for 2.0 perhaps, but mono was definitely too green then (as far as i can tell), so we didn't. The idea excites me like a kick to the head really.
--
|
<<attachment: zed-48.small.jpg>>
