[Kde-hardware-devel] inhibit suspend

[Dario Freddi]

Hello,

My mistake. I forgot to add a switch to the suspension lock handler. Sebas, 
the bug you're talking about should have been fixed since tonight. I'll fix 
suspension lock in a pair of days maximum.

Cheers
Dario

On Thursday 02 October 2008 15:43:41 you wrote:
> Le Thursday 02 October 2008, Sebastian Kügler a écrit :
> > I've played with the inhibit suspend feature last night. In fact, I've
> > already implemented it in the battery applet. I think the implementation
> > is wrong, though. The suspend should *only* be prevented when
> > powerdevil's idle timer times out, i.e. inhibit only affects the suspend
> > to save power, the machine going into suspend automatically after some
> > minutes. In fact, the suspend on critical battery level should always
> > ignore the inhibit. See it more like preventing an extreme screensaver
> > from kicking in.
> >
> > Invoking suspend 'by hand' still has to just work. In fact, if I click
> > the suspend button in kickoff or the battery applet, the machine should
> > release its cookies and just suspend, not prevent it -- that would annoy
> > a lot of users.
>
> Exactly. And that's exactly what the initial implementation I committed
> did. I admit I didn't take a close look at what Dario changes. I assumed it
> was keeping the previous behavior.
>
> As for "invoking suspend 'by hand'", we can still issue a warning to the
> user if there's some inhibits going on, probably showing it for 10s then
> proceeding in case the console is unattended. But we should never ever
> forbid the user to suspend if he wants to.
>
> > There also seems to be a bug somewhere. When powerdevil is running, I
> > don't seem to be able to suspend at all, I get the message "it's
> > inhibited". The user should always just override this setting.
> >
> > I'll seek to commit the inhibit feature to the battery later today.
>
> Excuse me, but what's the point of exposing that in the battery applet?
> It's merely a service provided to applications, it's not supposed to be
> triggered by the user.
>
> Regards.