On 15 Jul 2008, at 10:08, Vebjorn Ljosa wrote:
I had missed the fact that fullscreen_hook can be called even when the fullscreen frame parameterdoes not change (e.g., from x_set_font). I have cleaned/shortened yourfix a little; new patch is attached.
Yes, looks good to me.I was editing the patch directly with rather little oversight of the surrounding code...
This does not address the problem with certain (variable-width?) fontsand the frame not actually covering the whole screen,This problem should be fixed with the patch that Seiji checked in for meyesterday.
I assume that's the GNU Emacs CVS, right? Thanks
and neither does it disable frame movement by the toolbar.Since I'm on 10.4, I can't reproduce this. I don't think it's terriblyimportant.
I do, so I'll check how we can disable it.
A bigger problem, however, is that turning tool-bar-mode on and offwhile in fullscreen mode causes the frame to resize and become too shortor too tall. I will look into that.
iTerm turns off the toolbar for fullscreen mode. That would solve both problems.
It would be nice if this stuff could be fixed so the code can be checked into Emacs 23 before the feature freeze.I didn't understand from the previous discussion what kind of Mac support Emacs 23 has. Carbon? Cocoa?
Good point. Cocoa is getting merged this week. Carbon is dysfunctional in 23.Now that you mention it, I believe it's a waste of time to patch 23/ Carbon, but we should see how the patch can be adapted (or rewritten) for the Cocoa port. I'm not sure if the new port doesn't already support fullscreen.
Vebjorn, have you already signed papers with the FSF?No, but I'd be happy to. Are there instructions somewhere?
I'll look that up; I guess if we're not checking in your code, it's not a pressing matter.
- D
smime.p7s
Description: S/MIME cryptographic signature