Re: Would switching to Qt be a good idea?
samuel ammonius <[email protected]>
| Newsgroups | gmane.comp.desktop.xfce.devel.version4 |
|---|---|
| Message-ID | <CALbUA4j_6CtoEjiMEFFubhVyZPbk9x3vqa5E1=UGKtvuLcKUTQ@mail.gmail.com> |
> > "Switching to Qt" would basically mean completely rewriting everything > from scratch as far as I can tell. Not really. Most of XFCE's code is actually internal. Moving away from GLib > would mean rewriting everything from scratch, and since QtCore is actually > a replacement for GLib, it would be something that would need to be done > eventually. However, rewriting the graphical parts of XFCE to use another > toolkit won't be as difficult as it may sound. It also helps that Qt is > extremely well designed compared to GTK, even if we leave out the > gnomey-junk. Just look at this empty Qt button example > <https://riptutorial.com/qt/example/12519/hello-world> as opposed to this > empty GTK button example > <https://www.gtk.org/docs/getting-started/hello-world>. I could probably > rewrite a full copy of Mousepad alone in less than a week. On Fri, Jul 8, 2022 at 11:07 AM Mark Ballard <[email protected]> wrote: > I for one would welcome any effort to stop the over-simplification of > xfce. Operating Gnome apps is like trying to knit while wearing mittens, or > type with club fists. > > > On Fri, 8 Jul 2022 at 12:45, samuel ammonius <[email protected]> wrote: > >> Hi >> >> If you fork the project, do the work, attract enough developers to >>> maintain it, and eventually prove it's better, then it will become a >>> de-facto better project. Maintainers are not stubborn gatekeepers, they try >>> to do what's best for the project, but they rarely commit to accepting code >>> that has not been written yet. >>> >> >> Thanks for informing me about all of that. I'll try to start with smaller >> things like mousepad and the screenshooter, and move on to things like >> thunar and the panel when I'm familiar with the process of converting >> GTK/GLib to Qt. >> >> IMHO, time would be better invested in a Wayland compositor, but that's >>> another huge task. >>> >> >> What would a wayland compositor do? >> >> On Fri, Jul 8, 2022 at 4:27 AM Olivier Fourdan <[email protected]> wrote: >> >>> Hi >>> >>> On Thu, 7 Jul 2022 at 19:06, samuel ammonius <[email protected]> >>> wrote: >>> >>>> I posted a similar question on the XFCE forum, where a moderator told >>>> me I should go here. The forum post is here >>>> <https://forum.xfce.org/viewtopic.php?pid=67869#p67869>. >>>> >>>> The reason I was suggesting this is because GTK4 removed menus >>>> completely, because they were too "X11-centric". I think this is just an >>>> excuse to force people to use their designs, and XFCE's adoption of >>>> client-side decoration is proof that it's working. I'm not complaining >>>> about CSD in particular, but I'm trying to say that over the years, similar >>>> situations will arise and GTK will start to become a larger burden with >>>> every version that gets released. >>>> >>> I know that switching to Qt isn't something little. What I'm asking is, >>>> if I can fork all of XFCE's gui and make it use Qt, is it possible at all >>>> that it might get merged? >>>> >>> >>> This is a theoretical question impossible to answer without anything >>> tangible to evaluate. >>> >>> This is all open source, so eventually, the best ideas win. >>> >>> If you fork the project, do the work, attract enough developers to >>> maintain it, and eventually prove it's better, then it will become a >>> de-facto better project. Maintainers are not stubborn gatekeepers, they try >>> to do what's best for the project, but they rarely commit to accepting code >>> that has not been written yet. >>> >>> >>>> It won't take as long as it might sound because I've made both GTK and >>>> Qt applications and Qt is at least twice as easy to deal with. >>>> >>> >>> So if you believe it's a good idea, have the will, time and energy to do >>> it, then why not? >>> >>> IMHO, time would be better invested in a Wayland compositor, but that's >>> another huge task. >>> >>> Cheers >>> Olivier >>> _______________________________________________ >>> Xfce4-dev mailing list >>> [email protected] >>> https://mail.xfce.org/mailman/listinfo/xfce4-dev >> >> _______________________________________________ >> Xfce4-dev mailing list >> [email protected] >> https://mail.xfce.org/mailman/listinfo/xfce4-dev > > _______________________________________________ > Xfce4-dev mailing list > [email protected] > https://mail.xfce.org/mailman/listinfo/xfce4-dev _______________________________________________ Xfce4-dev mailing list [email protected] https://mail.xfce.org/mailman/listinfo/xfce4-dev