Gtk4 port progress as of 20230716

[sedwards]

This is just a checkpoint of what is needed to isolate the low-hanging fruit. The major changes would be getting away from gtk_container and moving to gtk_grid. I had an idea about how we could implement the gtk4 grid functions as inline wrappers for legacy gtk2/gtk3 but regardless, it will effectively mean a complete rewrite of the app to support gtk4.

We almost need to just start a new version. I would propose that since we are close to gtk3 support working, that we start tagging like gftp 2.99.1 and use the 2.99 series as the beta for the GTK3 release. Then once GTK3/Treeview support is done, we can finally tag a GFTP 3.0.x series with full GTK4 support.

Once that is done, we can start a new branch for the rewrite for GTK4 and we can use the 3.99.x tag to support it's development.

What do you think?

[wdlkmpx]

Yeah maybe 2.99 makes sense, but the gtktreeview transfer list introduces a new paradigm and there are bugs ( that have always been there) that need fixing (canceling inactive transfers, etc)

Other than that, gtk4 requires some huge changes, even porting to pure gtk3 will most likely result in a broken app that needs structural changes. I have seen the future

[sedwards]

Given how much is changing with your work on the Treeview, and then the cleanup of icons/xpm/png resources, maybe we should like tentatively plan on doing gFtp 3.0 with gtk3 support within say, 60 or 90 days. We don't have to rush things but it would be good to set a deadline, do some cleanup, and maybe get some life back in the project with a major revision, binary packages, PR, etc.

In the short-term I can keep looking for low-hanging fruit (I have a few more patches coming hopefully soon) and can setup some documentation tasks and cleanup.

I am going to leave this PR open for a while in case there is anyone watching that has interest in the status of GTK4 support...completely broken. But if someone wants to jump in and start hacking on cleanup to get ready for it, they can pull my patch. I'll remove the PR in a few weeks as we come up with a firm plan for the 3.x and 4.x lines of gFTP.