Hide X11 (Cmd-H) should trigger leaving fullscreen mode

[pelle@…]

While randr is very nice, it'd be even more great if it was possible to use Cmd-H to temporarily hide Xquartz. Currently, if I press Cmd-H while in randr mode, the screen will go black and nothing else happens. The only thing you can do after that is to press Cmd-Option-A to escape randr.

This is a regression from before randr captured the screen. Before that Cmd-H worked as expected.

Steps to reproduce: Open an xterm xrandr -s 1 Cmd-H

Expected result: Screen uncaptured Xquartz hidden.

Actual result: Black screen. No possible interaction except Cmd-Option-A or Cmd-Option-Escape.

Notes: It seems the Xquartz is hidden but still in charge of the captured display. Perhaps it'd be possible to simple uncapture the screen if a hide app event is received, and recapture it when reactivated.

[jeremyhu@…]

This is by design. You can override it with a preference.

From Xquartz(1):

defaults write org.macosforge.xquartz.X11 fullscreen_hotkeys -boolean true

Enable OSX hotkeys while in fullscreen

[pelle@…]

I think you misread it? The hotkey is activated. If it wasn't there wouldn't be a problem, as nothing would happen.

Try it and you'll see what I mean. :)

[jeremyhu@…]

So if you set:

defaults write org.macosforge.xquartz.X11 fullscreen_hotkeys -boolean false

... ?

[pelle@…]

The hotkey is activated regardless of that setting. I've tried both
defaults write org.macosforge.xquartz.X11 fullscreen_hotkeys -boolean true
and
defaults write org.macosforge.xquartz.X11 fullscreen_hotkeys -boolean false
and there's no change. But fixing it by disabling Cmd-H would at best be a workaround, IMO.

Sorry for the bad formatting in the steps to reproduce btw, I didn't realise it would remove those newlines. To make things clear, here's another version

• Quit XQuartz
• rm Library/Preferences/org.macosforge.xquartz.X11.plist
• open -a XQuartz
• (in the xterm) xrandr -s 1
• Click Allow
• Press Cmd-H

[jeremyhu@…]

Odd. Ok, I'll look into it for 2.7.0

[jeremyhu@…]

I have a fix for this which will be in my next PULL request to master. It is not in 2.7.1_rc2, but it will be in the next release.