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

List:       kfm-devel
Subject:    Re: kio::netaccess and imagepreview
From:       David Faure <david () mandrakesoft ! com>
Date:       2000-06-13 9:39:04
[Download RAW message or body]

On Tue, Jun 13, 2000 at 11:23:57AM +0200, Simon Hausmann wrote:
> 
> 
> On Tue, 13 Jun 2000, David Faure wrote:
> 
> > On Tue, Jun 13, 2000 at 11:06:12AM +0200, Simon Hausmann wrote:
> > > 
> > > (cc'ing to mosfet because of the thumbnails)
> > > 
> > > On Sun, 11 Jun 2000, David Faure wrote:
> > > 
> > > > On Sun, Jun 11, 2000 at 10:38:45AM +0200, Simon Hausmann wrote:
> > > > > The first one shows a nice loop of KIO::NetAccess::stat() being called
> > > > > recursively from local event loops (this seems a good candidate for
> > > > > Waldo's proposed reference counting?) .
> > > > 
> > > > Oooh. No, it's something different (the ref counting is in order to delay
> > > > the kapp->quit(), which is not the problem here).
> > > 
> > > Hmmm, seems I misunderstood the refcounting idea. I guess a reference
> > > counting which delays the kapp->quit() won't prevent the destruction of
> > > the KTMainWindow. Hmmmm, so we need some way to catch all close
> > > events? (installing an app-global event filter.. ouch ;-) no ;-)
> > 
> > I think I fixed this problem (of stat being called recursively
> > for the image preview thing), when I disabled the "active icon under mouse"
> > for thumbnails.
> 
> Ah, I see. But the user is still able to close the mainwindow during a
> "single" KIO::NetAccess::stat() or download() call, no? (because we still
> enter a "local" event loop)
> 
> And on slow systems like mine or when doing this over networks this
> happens quite easily.
> 
> Actually I guess every app using KIO::NetAccess and not doing a
> are-we-doing-a-net-access-operation check in MyKTMainWindow::queryClose is
> affected?

Sure. That's why we need to implement Waldo's refcounting solution,
so that the app doesn't exit as long as some download/stat is going on.
Is it a problem if the window is closed, though ? Hmmm... For NetAccess
it may be (since it's called from konqiconview). Maybe we also want a refcounting
per window (or even part), then ?

> > > > > BTW, shall we really name the thumbnail directory .mospics? I mean, it's
> > > > > probably cool for mosfet :-)) but it doesn't tell very much. How about
> > > > > calling it .kthumbs or so? Just my HO :-)
> > > > I agree. Now try to convince him :-) 
> > > 
> > > Mosfet? :-)
> > 
> > ;-)
> > And then he'll talk about compatibility with what the users already created :/
> 
> In my HO this is no problem, as we never said HEAD is meant to be
> compatible (at least on this level ;-) , until it is finally released.
> 
> What's your opinion?
Same as yours.

-- 
David FAURE
david@mandrakesoft.com, faure@kde.org
http://home.clara.net/faure/, http://www.konqueror.org/
KDE, Making The Future of Computing Available Today

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