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

List:       kde-devel
Subject:    Re: kded crashes my system (X)
From:       David Faure <david.faure () insa-lyon ! fr>
Date:       1999-06-27 20:54:17
[Download RAW message or body]

On Sun, Jun 27, 1999 at 10:47:46PM +0200, Simon Hausmann wrote:
> On Sun, 27 Jun 1999, David Faure wrote:
> 
> > This is the point where using a binary registry gets ... as bad as it is under
> > Windows. What about handling errors better in the registry loading, and if
> > anything fails, skip the loading and fall back on reading the ascii files
> > (which we do anyway - if modification time differs) ?
> > 
> > I know, easy to say, not to do (the "<<" operator doesn't really allow good
> > error checking !).
> 
> I fully agree with you.
> 
> Well, we can at least do some "minimum" handling by replacing the ASSERT
> calls (like in line 205 in kregfactories.cc) with proper error handling
> and read the registry in a "safe" way then (as you said it) , although I
> admit that it's no real solution.
Well, it's at least _part of_ the solution. We need to do that anyway.
BTW, it seems that this would take care of the bug reported above, wouldn't it ?
It fails in the assert, not in the '>>', right ?

> Hmmm... Is there any way to get rid of the operator way or to introduce
> error handling with it? Perhaps something like with the >>= operators of
> CORBA::Any ?
What does >>= do ?   - you have to get used to the fact that nobody understands
you when you talk about CORBA stuff, in this case not even me :)))

Perhaps istream has better methods than ">>", and which would be safer ?

-- 
David FAURE
david.faure@insa-lyon.fr, faure@kde.org
http://www.insa-lyon.fr/People/AEDI/dfaure/index.html 
KDE, Making The Future of Computing Available Today

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