>
> > > > I thought Magnus' changes were only in the current CVS branch, not
> > > > in REL6_4 ?
> > >
> > > You are absolutely right. Sorry Magnus. The COPY complaint I heard
> > > obviously was not from this.
> > >
> >
> > If we don't discover any errors on COPY FROM after some days,
> > I think I should fix it in REL6_4 too.
> >
> > Do we plan to release v6.4.3 sometimes? If so, are there any
> > neat things I could add to the v6.4.3 feature patch?
>
> You mean your fixes, right. Magnus's stuff was only in current, and is
> fixed now.
That one in COPY.
But the other one in ExecBRDeleteTriggers() is still in
place. If a trigger returns something it created with
SPI_copytuple() (what's done for PL/pgSQL triggers allways if
returning NEW or OLD), that heap_tuple isn't pfree()'d until
transaction end.
Since it's safe to throw it away I'll go ahead and put that
into REL6_4.
Jan
--
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me. #
#======================================== jwieck@debis.com (Jan Wieck) #
