[RFC Regression Fix] XWindowPeer override-redirect w/ metacity WM

[Andrew John Hughes]

On 16:22 Wed 14 Jul     , jon.vanalten at redhat.com wrote:
> 
> ----- "Andrew John Hughes" <ahughes at redhat.com> wrote:
> 
> > On 1 April 2010 20:47,  <jon.vanalten at redhat.com> wrote:
> > > Hi,
> > >
> > > It's been reported that applet windows that should be full-screen
> > are being displayed *behind* desktop elements.  See
> > https://bugzilla.redhat.com/show_bug.cgi?id=577953 for one such
> > report.  I agree with Deepak that this is a problem not limited to
> > applets, but within openjdk.  Afaict this is since
> > http://icedtea.classpath.org/hg/icedtea6/rev/83619682858e when
> > patches/icedtea-override-redirect-metacity.patch was removed.
> >  Changeset indicates the patch has been upstreamed in b17.  Closer
> > inspection reveals that only the first hunk is upstream, the second
> > part of the patch which actually addresses the metacity wm is not.  I
> > think this part of the patch needs to be restored.  Building locally
> > with that line back in XWindowPeer.java I see the correct behavior ie
> > desktop elements hidden by java window.  Diff attached.  Does anyone
> > see anything wrong with putting this back in?
> > >
> > > cheers,
> > >
> > > jon
> > 
> > 
> > See http://icedtea.classpath.org/wiki/IcedTea_JDK6_Patches
> > 
> > 'The Metacity issue has been marked as not a bug in the JDK, but one
> > in Metacity: see S6514512.'
> > 
> > http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6514512
> > https://bugzilla.gnome.org/show_bug.cgi?id=405269
> > 
> > No progress since 2007.  I think the better course of action would be
> > to at least ping the GNOME developers and find out what's going on. 
> > I
> > don't really want the patch back in IcedTea unless we have to,
> > especially when upstream have specifically objected to it.
> > -- 
> > Andrew :-)
> > 
> 
> Hate to flog the dead horse, but nobody's responded to the ping on the Gnome bug and with them originally taking the stance that they are complying with spec it seems unlikely that they will change metacity.  Maybe it is time for us to bring this patch back, so that users will have the "expected" behaviour?
> 

Ok, add it back on HEAD but make it very clear in the ChangeLog why
we're doing it with appropriate references to e-mails, etc.  The most
annoying thing about these sort of cases is that we have to do all the
research again to find out why we did such-and-such with a patch (as
I'm sure you've experienced too by now).

Thanks for keeping an eye on this.

> jon

Cheers,
-- 
Andrew :)

Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

Support Free Java!
Contribute to GNU Classpath and the OpenJDK
http://www.gnu.org/software/classpath
http://openjdk.java.net
PGP Key: 94EFD9D8 (http://subkeys.pgp.net)
Fingerprint = F8EF F1EA 401E 2E60 15FA  7927 142C 2591 94EF D9D8