[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-core-devel
Subject:    Re: DCOP without X server
From:       Stephan Kulow <coolo () kde ! org>
Date:       2001-12-02 11:13:44
[Download RAW message or body]

On Sunday, 2. December 2001 11:39, Allan Sandfeld Jensen wrote:
> On Friday 30 November 2001 15:49, David Faure wrote:
> > On Friday 30 November 2001 15:44, Matthias Ettrich wrote:
> > > On Friday 30 November 2001 15:26, Stephan Kulow wrote:
> > > > On Friday, 30. November 2001 15:16, David Faure wrote:
> > > > Yes, DCOP without dedicated DISPLAY is a session on it's own. E.g. a
> > > > daemon without X connection can't expect to give a process on the
> > > > user's desktop a command (at least I hope so). But e.g. a kioslave
> > > > should be useable in that session (even though for this case it may
> > > > be easier to develop from process forked slaves :)
> > >
> > > Agree. Who's going to do it?
> >
> > Me then.
> > Almost done.
> > Just not sure what to do in dcopserver_shutdown, since it already has
> > a fallback if DISPLAY isn't set, to ":0". Waldo, can you have a look ?
> > Only a problem if someone uses dcopserver_shutdown without a display,
> > though.
>
> One thing. Why limit the communication to one desktop in the first place?
> Remember it is the SAME user logged in two places, he might want them to
> cooperate. Imagine a command like move to desktop, with the extention of
> moving to another $DISPLAY.
> This would require a change of the DCOP protocol though, becouse
> applications only refering to the local desktop should add $DISPLAY as a
> parameter to the DCOP call.  But the potential is that two X-client could
> be made to behave like a dual monitor setup. :-)

First implement that, then we can still talk about extenting the DCOP 
protocol. I don't think you can move from display to display, but I may be 
wrong.

Greetings, Stephan

[prev in list] [next in list] [prev in thread] [next in thread]