KDE/kdebase/workspace/ksmserver

[Dario Freddi]

On domenica 08 febbraio 2009 19:00:36 Oswald Buddenhagen wrote:
> errrm ... how about using standard interfaces and hooking into session
> management instead? see the qt doc for starters.
>
> > and also, should I backport this?
>
> how about "no"? oh, wait, you *did* already backport ...

First of all, I'd like a more friendlier tone here. I am open to criticism, 
but not to this kind of weird talk. I have not committed anything that breaks 
stuff, and I've already read the Qt docs extensively, thanks.

There are other issues. If you look at my patch, I emit that signal right when 
logout() is called. Before it actually logs out, ksm triggers a session saving 
event, a knotify event to play the logout sound, etc. That's where the issue 
arise. If a suspend event is triggered in that very moment, there are no 
checks that allow one to know that the suspension should be avoided.

That signal simply notifies through DBus that a logout action has been 
triggered. Apart from that, if it is the signal that you don't like, I have 
already talked with Kevin and another possible solution could be triggering an 
inhibition right from KSMServer. Though a signal like that could be used by 
some other applications.

I think I specified this in the commit message: this signal has the purpose of 
notifying that the user is _about_ to log out. See the bug I was referring to 
for more infos.

If Qt session management already allows to catch those things, then it's my 
mistake and you can revert and refine my commit. In any case, I don't see a 
point in reacting like this, sorry.

Cheers,
Dario

-- 
-------------------

Dario Freddi
KDE Developer
GPG Key Signature: 511A9A3B

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/kde-core-devel/attachments/20090209/c2825149/attachment.sig>