Date: Wed, 13 Jun 2001 01:03:14 +0200 From: Palle Girgensohn <girgen@partitur.se> To: Akinori MUSHA <knu@iDaemons.org> Cc: ports@FreeBSD.org Subject: Re: cvs commit: ports/japanese/postgresql7 Makefile distinfo pkg-plist.doc ports/japanese/postgresql7/files patch-bd patch-jdbc1-statement Message-ID: <3B269FB2.5107225@partitur.se> References: <200106121445.f5CEjEe98188@freefall.freebsd.org> <863d95k8rd.wl@archon.local.idaemons.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Akinori MUSHA wrote:
>
> Hi,
>
> At Tue, 12 Jun 2001 07:45:14 -0700 (PDT),
> I wrote:
> > knu 2001/06/12 07:45:14 PDT
> >
> > Modified files:
> > japanese/postgresql7 Makefile distinfo pkg-plist.doc
> > Removed files:
> > japanese/postgresql7/files patch-bd patch-jdbc1-statement
> > Log:
> > Update to 7.1.2.
> >
> > PR: ports/28094
> > Submitted by: Saito Tomokatsu <saito@a2z.co.jp>
> >
> > Running scripts/configure.postgresql at the post-extract stage is too
> > late. Do it at pre-everything to reflect added BUILD_DEPENDS entries.
>
> Since BUILD_DEPENDS is used on fetch, creating Makefile.inc at
> pre-extract is too late. We have a target called "pre-everything" for
> this kind of situation.
>
> I think databases/postgresql7's Makefile also needs this change. See
> the attached patch and let me know if it works all right.
Correct. Please see PR 27949. I get a lot of questions about
this, and would really appreciate if it could be comitted. I
initially used pre-fetch, but this is because I didn't know
about pre-everything. Most other ports using dialog(1) scripts
seem to create the Makefile.inc in the pre-fetch phase, and
this is what I did as well. Then some helpful comitter changed
this without telling me first, just because he liked to be able
to do make fetch-depend overnight, and not have the fetch stop
halfway to run a configure script. This should of course be
done with make -DBATCH fetch-depend... His intentions where
good, but introduced a bug. It happens all the time ;-)
pre-fetch or pre-everything, not much difference I guess.
Please feel free to commit my POR 27949 with the configure
script run in either target. It is probably best to keep in
sync with the japanese port, IMO.
> (To see the problem, try building the port with the JDBC option turned
> on when jakarta-ant is not installed.)
or tcl/tk...
Regards,
Palle
>
> --
> /
> /__ __ Akinori.org / MUSHA.org
> / ) ) ) ) / FreeBSD.org / Ruby-lang.org
> Akinori MUSHA aka / (_ / ( (__( @ iDaemons.org / and.or.jp
>
> "Freeze this moment a little bit longer, make each impression
> a little bit stronger.. Experience slips away -- Time stand still"
>
> Index: Makefile
> ===================================================================
> RCS file: /home/ncvs/ports/databases/postgresql7/Makefile,v
> retrieving revision 1.85
> diff -u -u -r1.85 Makefile
> --- Makefile 2001/06/04 11:52:52 1.85
> +++ Makefile 2001/06/12 15:34:59
> @@ -89,7 +89,7 @@
> NO_OPENSSL="${NO_OPENSSL}" \
> DISTNAME="${DISTNAME}"
>
> -pre-extract:
> +pre-everything:
> @ ${SETENV} ${SCRIPTS_ENV} ${SH} ${SCRIPTDIR}/configure.postgresql
>
> # We must .include here because we need the Makefile.inc @ pre-install,
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3B269FB2.5107225>