Jean-Marc Lasgouttes a écrit :
"Lars" == Lars Gullik Bjønnes <[EMAIL PROTECTED]> writes:
Lars> Note that with subversion (yes, I have my scripts ready and
Lars> warmed up for conversion), moving things around in the tree or
Lars> duplicating them is a breeze. So unless the build system is
Lars> re-vamped in the first seconds after 1.4 branch/release this is
Lars> not necessary, and imho the wrong focus.
You are right. Actually, I would ask for something different now:
since it is obvious that the qt4 frontend is going to have large
surgery in 1.5, there is little point to start from a common base in
1.4. What I would like in 1.4 is a qt4 frontend which is as close as
reasonable to the qt2 frontend. That would mean no big refactorization
of code, optimizations, whatever. I do not care if there are still
some Qt3 compatibility classes in there, for example.
Abdel, does it seem reasonable? Once qt4 is officially in the 1.4
tree, I am going to be conservative about further changes anyway. And
before you ask, it will be the same for gtk :)
Then, qt4 would be 1.5 only because my port has a lot of refactorization
already and I am not willing to go backward. That's fine with me, the
qt2 frontend is good enough for 1.4 and I don't see the benefits of
having a "qt2 alike" qt4 frontend. As the qt2 frontend seems "frozen"
for the last year or so, I do not think developers would shout if we say
that all new Qt functionality should be qt4 only ;-)
Abdel.
JMarc
