On 7/14/21 7:41 AM, Chris Down wrote:
Hey folks,Laslo Hunhold writes:count me in in that regard. If an application (most likely a game) wants exclusive fullscreen, it can capture the mouse in the window. I always set it like this in wine and have had no problems with that, and it still allows workspace-switching. For what it's worth, in my humble opinion dwm should always guarantee that you can switch workspaces. "Exclusive" fullscreen is a hack as we know from slock.Probably worth me chiming in as the commit author :-)In my case, a tangible problem exists for cases where, for example, mpv is playing a video full screen. If one then accidentally performs a focusstack (as I have many times) then they cannot possibly see which window is being selected, and it's somewhat cumbersome to recover focus to the foreground in order to regain control.
Would you mind describing how you accidentally perform a focusstack? What keybindings do you have it set to that makes this so easy to do without intending to? Or do you mean that you intentionally press the focusstack keybindings, but forget that a window is in fullscreen?
Looking at the conversation, it seems like the concern is about interaction with the fakefullscreen[0] patch, if I understand correctly.That is how I came across this issue, personally. But to me, the bigger issue is that there is *no reason* for this behavior to exist except the assumption that a user's explicit inputs *must* be accidental and therefore should be ignored. This is unintuitive and undocumented, and just a Bad Idea that dwm shouldn't implement. But I digress.
As I understand it, the behaviour in mainline makes sense with stock dwm, just not with this patch. If that's the case, can't one just change the behaviour as part of the fakefullscreen patch on the wiki? Or maybe I'm misunderstanding the use case in question?0: https://dwm.suckless.org/patches/fakefullscreen/
I have now just done that. I sent a patch updating fakefullscreen to use the new lockfullscreen variable introduced in this patch. I didn't realize this, but there was another version of fakefullscreen added in May that reverted your commit, which is the effective behavior introduced by this patch. However, it seems this option is supported by others, even those who may not use fakefullscreen, so it is good that it is in mainline now, if your patch is to be mainline behavior as well.
--- Sebastian LaVine | https://smlavine.com
OpenPGP_0x819C7D054C7C1465.asc
Description: OpenPGP public key
OpenPGP_signature
Description: OpenPGP digital signature
