On 20.02.2012, at 23:25, Bill Spitzak wrote: > Is the design of the "wrappers" in fltk3.0 intended to allow the > widget's to be *completely* different, in that they can have different > sets of virtual functions (such as changing the arguments, removing > some, etc), different hierarchies of objects, and reuse names from 1.0 > for different purposes (like making x() be relative, while the 1.0 > wrapper still returns absolute)?
Yes, exactly. I am hoping that old FLTK1 apps will just run under FLTK3. > I'm going to assume this is true for any further discussion. There > should now be an ability to, if wanted, completely rewrite all the > widgets. This may offer a lot more freedom than 1.0. > > The biggest question is exactly what to do if such rewriting is > attempted. Is there a plan to stop contributions to the fltk1 branch, or > to carefully reproduce any and all fixes in fltk3, even if the base code > diverges a lot? The loss of all fixes to fltk1 in fltk2 is what killed > it, apparently. I would love to get to the point - rather sooner then later (but then again, I was inactive for half a year). At some point when the emulation layer works and FLTK 1.3 is stable, we should stop contributing to it and only fix catastrophic bugs. All existing users can then migrate to FLTK3, even mixing FLTK 1 and 3 widgets and coding, so existing widgets are not lost. But who knows. I assume that some huge advantage of FLTK 3 will make all users *want* to migrate. Maybe the compositing engine that we discussed earlier could be that reason? We would finally get much nicer design for all widgets... . _______________________________________________ fltk-dev mailing list [email protected] http://lists.easysw.com/mailman/listinfo/fltk-dev
