Thanks Steve. Applied.
I hope to make a new DBI release within a few days.
Tim.
On Mon, Oct 23, 2006 at 09:10:38AM -0500, Steve Peters wrote:
> On Sat, Oct 14, 2006 at 08:40:14AM +0200, Andreas J. Koenig wrote:
> > >>>>> On Fri, 13 Oct 2006 17:00:46 +0200, demerphq <demerphq [at] gmail.com> said:
> >
> > > Are there any other issues not listed above?
> >
> > Showstopper is that DBI doesn't compile. There seems to be no RT
> > ticket but xray has some hits:
> >
> > http://www.xray.mpe.mpg.de/cgi-bin/w3glimpse/perl5-porters?q uery=breaks+dbi&errors=0&case=on&maxfiles=100&maxlines=30
> >
>
> I seemed to get it to compile, but it blew up on assertions on assertions.
> With some help from Nicholas Clark, the following patch gets DBI-1.52 to
> compile *and* pass all of its tests.
>
> Steve Peters
> steve [at] fisharerojo.org
>
> --- DBI.xs.old 2006-10-23 09:08:08.000000000 -0500
> +++ DBI.xs 2006-10-23 09:02:50.000000000 -0500
> [at] [at] -2615,7 +2615,7 [at] [at]
> */
> if (SvROK(h) && SvRMAGICAL(SvRV(h)) && (mg=mg_find(SvRV(h),'P'))!=NULL) {
>
> - if (mg->mg_obj==NULL || !SvOK(mg->mg_obj) || SvPVX(mg->mg_obj)==NULL) { /* maybe global destruction */
> + if (mg->mg_obj==NULL || !SvOK(mg->mg_obj) || SvRV(mg->mg_obj)==NULL) { /* maybe global destruction */
> if (trace_level >= 3)
> PerlIO_printf(DBILOGFP,
> "%c <> %s for %s ignored (inner handle gone)\n",
>
