-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Monday 12 January 2004 19:08, Lubos Lunak wrote:
> On Monday 12 of January 2004 17:33, George Staikos wrote:
> > On Monday 12 January 2004 11:24, Lubos Lunak wrote:
> > > > >  That's not going to be only khtml, that's going to be everything
> > > > > using kwallet. I'm not sure if people would like to always have to
> > > > > open wallets asynchronously.
> > > >
> > > >   Exactly, reverting callExt for all applications will probably break
> > > > many of them.
> > >
> > >  What do you expect to break, besides the popup lockup? I was only
> > > suggesting that people might not like API-wise being forced to open the
> > > wallet asynchronously, I don't see any potential problem with making
> > > the call not to use event loop and only somehow handle the popup
> > > problem.
> >
> >    The GUI stops updating while it blocks in the DCOP call in synchronous
> > mode without using callExt.
>
>  I'm afraid GUI not updating is insignificant when compared to crashes or
> lockups. Code in libraries should IMHO never re-enter the event loop,
> because you never know what the event loop might do in the middle of
> something that's considered to be "atomic" by the developer calling it.

Can this problem be related to bug #71419 (and see my mail on kde-core-devel: 
"HELP needed: NetAccess::stat blockes when switching desktop")? There the 
event loop used in NetAccess is not exited after desktop switching.

Andras

- -- 
Quanta Plus developer - http://quanta.sourceforge.net
K Desktop Environment - http://www.kde.org
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.2-rc1-SuSE (GNU/Linux)

iD8DBQFAAt2kTQdfac6L/08RAjokAKCRQltJUcmp1wqxz8FUKnQbjRPDSwCffxfg
/nNWPBJMxXQ4YGZIH28ihl4=
=5IQj
-----END PGP SIGNATURE-----