Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 27 Feb 2002 03:13:24 -0300
From:      Mario Sergio Fujikawa Ferreira <lioux@FreeBSD.org>
To:        Maxim Sobolev <sobomax@FreeBSD.ORG>
Cc:        Will Andrews <will@csociety.org>, portmgr@FreeBSD.ORG, FreeBSD-ports@FreeBSD.ORG, Patrick Li <pat@databits.net>, Pete Fritchman <petef@databits.net>
Subject:   Re: bsd.port.mk MASTER_SITES_NN misbehavior (was ghostscript-afpt and MASTER_SITES_NN patch)
Message-ID:  <20020227061346.13502.qmail@exxodus.fedaykin.here>
In-Reply-To: <3C769D92.7C1C441F@FreeBSD.org>; from sobomax@FreeBSD.ORG on Fri, Feb 22, 2002 at 09:35:24PM %2B0200
References:  <20020221221039.19324.qmail@exxodus.fedaykin.here> <3C764483.F20F486C@FreeBSD.org> <20020222144830.GS82673@squall.waterspout.com> <3C769D92.7C1C441F@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help

--cWoXeonUoKmBZSoM
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Hi,

	After lots of 'make -d A' which is not the most fun
way to spend your nights.... :)
	I've got a fix for the recursiveness problem. Actually, the
final code was designed to not contain this flaw. I must have
confused my versions submitting an old one when I sent it to Will
Andrews for inclusion on the 4-exp run. :( Bad bad bad lioux. Sorry
Will.
	Anyway, this patch brings it back to what I had in mind at
first.
	The patch is available at

http://people.freebsd.org/~lioux/patch-bsd.port.mk

	Patches to test it demonstrating its usefullness can be
found at

- for ghostscript-afpl
http://people.freebsd.org/~lioux/patch-print:ghostscript-afpl:Makefile

- for qmail
http://people.freebsd.org/~lioux/patch-mail:qmail:Makefile

	However, one interesting problem arised. The following patch
shows a problem

- for ghostscript-afpl
http://people.freebsd.org/~lioux/patch-print:ghostscript-afpl:Makefile-brok=
en

	However, the difference between the working and the -broken
version is that working uses group GIMP_PRINT whilst -broken uses
GIMP-PRINT.  I make no assumptions about group naming. Furthermore
- is a special character as far as my patch is concerned.

diff patch-print:ghostscript-afpl:Makefile patch-print:ghostscript-afpl:Mak=
efile-broken    =20
15c15
< +MASTER_SITES=3D        ${MASTER_SITE_SOURCEFORGE:S/$/:DEFAULT,GIMP_PRINT=
,HPINKJET/} \
---
> +MASTER_SITES=3D        ${MASTER_SITE_SOURCEFORGE:S/$/:DEFAULT,GIMP-PRINT=
,HPINKJET/} \
22c22
< +MASTER_SITE_SUBDIR=3D  ${PORTNAME} gimp-print/:GIMP_PRINT hpinkjet/:HPIN=
KJET
---
> +MASTER_SITE_SUBDIR=3D  ${PORTNAME} gimp-print/:GIMP-PRINT hpinkjet/:HPIN=
KJET
31c31
< +             ${HP970_DRV}:HP970_DRV ${GPRINT_SRC}:GIMP_PRINT \
---
> +             ${HP970_DRV}:HP970_DRV ${GPRINT_SRC}:GIMP-PRINT \

	I wonder. Is this a make bug? What can we do to overcome it?
Can people reproduce this? Track it down?
	If this "bug" is not related to my patch but to make itself,
then we only have one outstanding bug (mentioned below) to call
MASTER_SITES_NN prime. :-)

	For reasons beyond my limitted make(1) knowledge, the following
code does not stop any port from using a group of all, ALL or
default. These words NEED to be reserved for now since they are
meant to have special meaning: I will be introducing code for that
later

=2E                       if ${_group}=3D=3Dall || ${_group}=3D=3DALL || ${=
_group}=3D=3Ddefault
=2E                       BEGIN:
                                @${ECHO} "The words all, ALL and default ar=
e reserved and cannot be used"
                                @${ECHO} "in group definitions."
                                @${ECHO} "Please fix your MASTER_SITES"
                                @${FALSE}
=2E                       endif

Check similar non-working code at bsd.port.mk lines 1259, 1276,
1298 and 1319. I really NEED this to work.

	Please, comment on it. I believe portmgr will be including
this in 4-exp run on this weekend (or the next) so I need comments
on the problems (to fix them before it is tested on bento, or so
that it CAN be tested on it).

	Regards,

--
Mario S F Ferreira - DF - Brazil - "I guess this is a signature."
Computer Science Undergraduate | FreeBSD Committer | CS Developer
flames to beloved devnull@someotherworldbeloworabove.org
feature, n: a documented bug | bug, n: an undocumented feature

--cWoXeonUoKmBZSoM
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (FreeBSD)
Comment: For info see http://www.gnupg.org

iD8DBQE8fHkZXvSymrg2XlURAvYIAJ9dDYw8iAAOymOPo+IS19pK68xLGwCfQjPQ
yF8hJqwru7IxL1ymgpGLHXk=
=jaZn
-----END PGP SIGNATURE-----

--cWoXeonUoKmBZSoM--

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?20020227061346.13502.qmail>