Bug 413964 - Make rsibreak's mouse movement detection less sensitive.
Summary: Make rsibreak's mouse movement detection less sensitive.
Status: REPORTED
Alias: None
Product: rsibreak
Classification: Applications
Component: general (show other bugs)
Version: 0.12.8
Platform: Arch Linux Linux
: NOR wishlist
Target Milestone: ---
Assignee: Albert Astals Cid
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2019-11-09 01:02 UTC by poperigby
Modified: 2022-03-08 11:32 UTC (History)
3 users (show)

See Also:
Latest Commit:
Version Fixed In:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description poperigby 2019-11-09 01:02:36 UTC
Simply bumping the table your mouse is on will reset the timer. I think the sensitivity should be cranked down quite a bit. The timer should only reset if the mouse is move an inch or so.
Comment 1 Patrick 2020-11-24 22:36:14 UTC
I am also running into this issue, but its a bit more severe. For some reason my mouse (USB) always sets off the activity timer even when not moving. It doesn't stop by machine from sleeping and I have confirmed its the mouse by unplugging it during the countdown. I have been able to workaround this by un-checking "flash on activity", but it would be nice to be able to have more control here (like dialing in the sensitivity as poperigby@mailbox.org suggests and/or disabled activity tracking (while keeping keyboard) all together. Possibly by settings mouse sensitivity to zero.
Comment 2 Oswald Buddenhagen 2022-03-08 11:32:59 UTC
that issue is more or less unsolvable at the rsibreak level: it uses the kidletime library, which uses whatever the windowing system reports to it. so basically, you need to file feature requests against xorg/xserver and potentially numerous wayland compositors.

rsibreak can't track the events itself, as security measures against key-logging (are supposed to) prevent that. though it might be possible to configure a special authorization - i didn't check what options the display servers offer nowadays. that would also have the side effect of potentially fixing #392679.