QLabel kills our KMessageBoxes (and others)
Lubos Lunak
l.lunak at suse.cz
Fri Oct 19 10:12:04 BST 2007
On Thursday 18 of October 2007, Thomas Zander wrote:
> On Thursday 18 October 2007 13:09:19 Lubos Lunak wrote:
> > Nah. The "unfixable" part is the fact that with multiline lineedits
> > there's no obvious minimum size (since you can stretch the width or
> > height while reducing the other and the contents will still fit) and
> > "can't be fixed in X11" means that the window manager spec has support
> > for minimum size, but only for fixed minimum size and not any
> > width-for-height size.
> >
> > TT apparently are not willing to do anything besides a perfect fix
> > (which is not possible as they say), but I think e.g. going with a
> > reasonable ratio of width/height as the minimum size could work good
> > enough for most cases.
>
> Speaking as a co-KDE guy; I know you are well known in the window-manager
> area so I'm wondering what is holding up the 'perfect fix' ?
> Can you perhaps work in the spec to make this a concept available on X11?
I'm not sure. With fixed-value minimum size it's simple, both to set and to
use. But with height_for_width minimum size, with more complex layouts it's
probably possible to end up with more complicated rule than something
like "height = 20 + width / 50". Now, the window manager should never be
blocked by a client, meaning it can't query the client whenever it needs this
value - so how should it technically get the minimum size?
The best thing I can think of is running a loop over all width values and
when showing a window giving the WM a list of (width,height) pairs for all of
these. Just trying to find a decent fixed minimum size as I've suggested
above seems a better try.
--
Lubos Lunak
KDE developer
--------------------------------------------------------------
SUSE LINUX, s.r.o. e-mail: l.lunak at suse.cz , l.lunak at kde.org
Lihovarska 1060/12 tel: +420 284 028 972
190 00 Prague 9 fax: +420 284 028 951
Czech Republic http//www.suse.cz
More information about the kde-core-devel
mailing list