On Fri, Feb 28, 2003 at 05:30:59PM +0100, Dominik Vogt wrote:
> On Fri, Feb 28, 2003 at 04:27:15PM +0100, Olivier Chapuis wrote:
> > On Thu, Feb 27, 2003 at 12:06:42PM -0500, Tessa Lau wrote:
> > > On Thursday, Olivier Chapuis mumbled:
> > > > It seems that Dan and me do not have this problem (without the
> > > > FixedPPosition style). Can you send me a minimal java program
> > > > which has this problem.
> > >
> > > Certainly. *Every* Java Swing program I run is displayed off the
> > > upper-left corner of the screen, but I'll attach a minimal program for
> > > you to play with. ("javac Misbehave.java; java Misbehave")
> > >
> > > I suspect it's a windowmanager/JDK interaction. I'm running FVWM
> > > 2.5.5. I've tried running this program with four different JDKs, and
> > > the bug is only triggered by 1.4.1:
> > >
> > > IBMs Java 1.3.1 works correctly
> > > Blackdown Java 1.3.1-02b-FCS works correctly
> > > Blackdown Java 1.4.1-beta BUG
> > > Blackdown Java 1.4.1-01 BUG
> > >
> >
> > And
> >
> > SUN JDK 1.3.1_04 "works correctly" (here)
> > Sun JDK 1.4.1-01 "works correctly" (Dan)
> > Bob jdk BUG (Bob want jdk do you use?).
> >
> > I say "works correctly" because I do not understand why when you
> > do not set any location java assumes that the location is (0,0).
> > I imagine this is intentional.
> >
> > > Hope that helps you to track down the problem.
> > >
> >
> > Can you send "a bug report" to Blackdown? Do you try with others
> > window managers?
> >
> > If I well understand this depends on how the application think the wm
> > will interpret a move-resize config request. Maybe, we should add a
> > style to handle such situations (I do not think that FixedPPosition is
> > really appropriate). Dominik, what do you think?
>
> I tried several times to discuss that topic on the wm-spec list,
> but all that ever comes out of this is that people do not care and
> say "but the ICCCM is clear about this". They seem to completely
> ignore that the problem has been there for a long time and can
> only be fixed with a clear spec.
>
Yes. Today I reread the threads on this subject on the wm-spec list.
I am totally agree with you. The problem is that a lot of applications
use now the ICCCM way for move-resize after mapping: some jdk (but the
pb is more complex here), gtk and tk (and we got a lot of bug report
on tk related to this pb). I do not know for qt/kde. On the other
hand, I am afraid that nothing will be done on this in the EWMH spec
(basically because havoc/metacity does not care about legacy app which
use the gravity).
So, I really think that we need a style ICCCMMoveResize / !ICCCMMoveResize
The default is not really the pb I think. In any case we will have some
bug reports on this. With such a style we will have a good answer.
About implementation, I think I will need a lot of time and as you
know perfectly the pb you can do it in a few minutes ...
Moreover, feature freeze is not a pb here as this style will close
some "important bugs".
Olivier
PS: I've also read your mail on SUN Bug Parade (bug nbr. 4401846)
I like it a lot. Tessa, if you want a clear explanation on the pb
take a look at it (java SUN site -> bug parade -> register -> search
4401846)
--
Visit the official FVWM web page at <URL:http://www.fvwm.org/>.
To unsubscribe from the list, send "unsubscribe fvwm-workers" in the
body of a message to [EMAIL PROTECTED]
To report problems, send mail to [EMAIL PROTECTED]
