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

List:       kde-core-devel
Subject:    Re: This Week's CVS issues :)
From:       Michael Brade <Michael.Brade () informatik ! uni-muenchen ! de>
Date:       2001-06-02 9:10:19
[Download RAW message or body]

On Saturday 02 June 2001 00:44, David Faure wrote:
> On Saturday 02 June 2001 00:04, Michael Brade wrote:
> > Oops, thanks for the hint and the fix :) But I think to add an emit
> > completed() (and emit started() as well) would be the correct fix.
> > However, a problem could be that an app might do "KDirLister::job()"
> > without testing for the returned job being null. What do you think/should
> > I commit?
>
> I don't have the big picture anymore... is it sensible that KDirLister
> emits completed() without having emitted started() before ? 
No, not really. But it is cleaner to do so - IMHO.

> Does it emit
> started() for automatic-updates ?
Yes, of course.

> Hiding/showing the dot files looks like
> an automatic update (it actually uses the same code if showing them, which
> is sensible). I'm not sure that emitting completed() in this case (which is
> manually triggered by the application anyway, unlike the usual async stuff)
> is necessary/wanted.
Hmm, the problem with not emitting completed() is that you will never know 
when the last file to delete is emitted (KDirLister emits all at once ATM but 
you can't and shouldn't rely on this). Even more, showing already hidden 
files _will_ emit started() and completed(). So I'd definitely vote for 
emitting the signals in the hide-dotfiles-case as well, especially after 
thinking about this again with enough sleep ;-)


Ciao,
  Michael

-- 

       Some operating systems are called `user friendly',
             Linux however is `expert friendly'.

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