Friso Smit

Results 322 comments of Friso Smit

Idea: Don't close actions immediately upon a notification timing out

But I don't know how other programs implement their action code. It would be useful to know how many programs can handle waiting for an action indefinitely without problems

dunst shows the same icon when the content of --icon is changed

You mean the timestamp of when the icon was changed on disk? Adding that timestamp should fix this issue yes

dunst shows the same icon when the content of --icon is changed

Yeah sounds good

dunst shows the same icon when the content of --icon is changed

Could be. It probably was used before. Iirc it contained the hash of the icon data.

A ghost rectangle is rendered for a split second before the actual notification balloon

I still don't know what causes the issue, so it's hard to fix. The commit suggested by pvonmoradi doesn't seem to introduce the issue, but potentially make it stand out...

Enter infinite loop when surface is destroyed by compositor

I think it can technically still happen. But it should not happen if the compositor behaves normally

Reload config by SIGUSR1

Please don't bump issues. If you want that it is implemented you can take a look at my proof of concept (#968) and try to improve on the bugs it...

Feature: Add negative comparison for regex (and other things?)

Could be feasible yeah. You'll have to see if it's worth the effort

Wayland: detect the right display even when follow=keyboard/mouse

> but now it's whatever monitor I was on when the notification was sent. That's the intended behavior. If the notification doesn't follow your cursor, then you should check with...

Wayland: detect the right display even when follow=keyboard/mouse

You can try older versions of sway to see if it works there. I don't think dunst has changed anything here. This isssue is also about something different