[RFC][icedtea6]: PR677 fix

[Denis Lila]

> Are you applying these in the same order they were applied to
> OpenJDK7?

Yes.

> It worries me that you're adding these patches to the top of the patch
> list and not the end. Is there a good reason for this?

The reason is to apply the patches in the same order
in which they went into the OpenJDK7 trees. If I added
them to the end there would be patches before them that
would make them not apply (6693253-security_warning.patch,
in particular). By putting them in chronological order they
instead break 6693253-security_warning.patch, but this
is easy to fix: just replace the problematic hunks with
the hunks from the openjdk7 changeset.

If I put the new patches at the end, I would have to modify
the new patches to make them apply, and these modifications
would be much more complicated.

Should I go ahead and push the patch?

Thank you,
Denis.

----- Original Message -----
> On 17:29 Mon 06 Jun , Denis Lila wrote:
> > Hi.
> >
> > The fix for this is a backport. I've attached two
> > patches. Each of them fixes it. The first is much
> > simpler, but it is not a complete backport of the
> > fix: it only takes the changes in Window.java, but
> > not src/windows/native/sun/windows/awt_Dialog.{cpp,h}.
> >
> > ChangeLog (for pr677-fix-simple.patch):
> > +2011-06-06 Denis Lila <dlila at redhat.com>
> > +
> > + * patches/openjdk/6769607-modal-hangs.patch:
> > + New patch. Fixes PR677.
> > + * Makefile.am: Apply patch.
> > + * NEWS: Update with backport.
> > +
> >
> >
> > The second patch is a complete backport, but it's much
> > bigger. I had to backport a couple of other changesets
> > to make it apply, and I also had to tweak an existing
> > patch.
> >
> > ChangeLog (for pr677-fix-complete.patch):
> > +2011-06-06 Denis Lila <dlila at redhat.com>
> > +
> > + * Makefile.am: Apply patches.
> > + * NEWS: Update with backports.
> > + * patches/openjdk/6578583-modality-broken-vista.patch:
> > + * patches/openjdk/6610244-modal-fatal-error-windows.patch:
> > + * patches/openjdk/6769607-modal-hangs.patch:
> > + New patches. The last fixes PR677. The other two are
> > + necessary for the last to fully apply.
> > + * patches/openjdk/6693253-security_warning.patch:
> > + Replace the awt_Dialog.cpp hunk with the corresponding hunk
> > + from the OpenJDK7 changeset of which this patch is a backport.
> > + Without this change, this patch doesn't apply unless the
> > + previous 3 are removed.
> > +
> >
> >
> > I favour the first patch because with it we don't have to modify
> > 6693253-security_warning.patch, but if it's important that
> > we keep the source changes as close to ojdk7's as
> > possible (even for windows only files), we can go with
> > the second.
> >
> > Which one should I push?
> >
> > Regards,
> > Denis.
> 
> The second is the right choice. The first adds an untracable hunk of
> code which claims to be a backport of a bug, but isn't.
> 


> --
> Andrew :)
> 
> Free Java Software Engineer
> Red Hat, Inc. (http://www.redhat.com)
> 
> Support Free Java!
> Contribute to GNU Classpath and IcedTea
> http://www.gnu.org/software/classpath
> http://icedtea.classpath.org
> PGP Key: F5862A37 (https://keys.indymedia.org/)
> Fingerprint = EA30 D855 D50F 90CD F54D 0698 0713 C3ED F586 2A37