svn commit: r518052 - head/sysutils/am-utils

[Mathieu Arnold]

On Thu, Nov 21, 2019 at 05:23:33AM -0800, Cy Schubert wrote:
> In message <20191121071847.6564ybobj4ennhcy at atuin.in.mat.cc>, Mathieu 
> Arnold wr
> ites:
> > 
> >
> > --wcbpoteyztvskbdz
> > Content-Type: text/plain; charset=us-ascii
> > Content-Disposition: inline
> > Content-Transfer-Encoding: quoted-printable
> >
> > On Thu, Nov 21, 2019 at 08:09:24AM +0100, Tobias Kortkamp wrote:
> > > On Wed, Nov 20, 2019 at 09:29:44PM +0000, Cy Schubert wrote:
> > > > Author: cy
> > > > Date: Wed Nov 20 21:29:44 2019
> > > > New Revision: 518052
> > > > URL: https://svnweb.freebsd.org/changeset/ports/518052
> > > >=20
> > > > Log:
> > > >   Make the installation of the rc script conditional to avoid conflict
> > > >   with base.
> > > >  =20
> > > >   MFH:		2019Q4
> > > > [...]
> > > > -USE_RC_SUBR=3D	amd
> > > > +RC_USE_RC_SUBR=3D	amd
> > >=20
> > > There is no such options helper.  It should be
> > >=20
> > > RC_USE=3D		RC_SUBR=3Damd
> > >=20
> > > Also the rc script seems to still use /usr/sbin/amd instead of
> > > %%PREFIX%%/sbin/amd which should probably be fixed too.
> >
> > Also, I am unsure of the use of this rc script, if it has the same name
> > as the base system one, they will have the same variables controlling
> > them, and both will be started at boot time if amd_enable=3Dyes, which is
> > probably not what is wanted here.
> 
> It does install correctly. A partial list of sysutils/ ports, including 
> some of mine.
> 
> /opt/src/new-ports/sysutils/nut/Makefile
> 44:USE_RC_SUBR=	nut nut_upsmon nut_upslog
> 
> /opt/src/new-ports/sysutils/openupsd/Makefile
> 13:USE_RC_SUBR=	${PORTNAME}
> 
> /opt/src/new-ports/sysutils/afbinit/Makefile
> 24:USE_RC_SUBR=		afbinit
> 
> /opt/src/new-ports/sysutils/graylog/Makefile
> 25:USE_RC_SUBR=	graylog
> 
> /opt/src/new-ports/sysutils/bacula9-client/Makefile
> 9:USE_RC_SUBR=	bacula-fd
> 
> /opt/src/new-ports/sysutils/fcron/Makefile
> 31:USE_RC_SUBR=	fcron
> 
> /opt/src/new-ports/sysutils/linux-crashplan/Makefile
> 21:USE_RC_SUBR=	crashplan
> 
> /opt/src/new-ports/sysutils/riak-cs/Makefile
> 22:USE_RC_SUBR=	riak_cs
> 
> /opt/src/new-ports/sysutils/cpupdate/Makefile
> 22:USE_RC_SUBR=	${PORTNAME}
> 
> /opt/src/new-ports/sysutils/syslog-ng321/Makefile
> 51:USE_RC_SUBR=	syslog-ng
> 
> /opt/src/new-ports/sysutils/fluent-bit/Makefile
> 20:USE_RC_SUBR=	${PORTNAME}
> 
> /opt/src/new-ports/sysutils/consul-alerts/Makefile
> 26:USE_RC_SUBR=	${PORTNAME}
> 
> /opt/src/new-ports/sysutils/diskcheckd/Makefile
> 16:USE_RC_SUBR=	diskcheckd
> 
> /opt/src/new-ports/sysutils/asusoled/Makefile
> 17:USE_RC_SUBR=	asusoled oled_clock
> 
> /opt/src/new-ports/sysutils/pprotectd/Makefile
> 12:USE_RC_SUBR=	${PORTNAME}
> 
> /opt/src/new-ports/sysutils/runit/Makefile
> 20:USE_RC_SUBR=	runsvdir
> 
> /opt/src/new-ports/sysutils/monitord/Makefile
> 13:USE_RC_SUBR=	monitord
> 
> /opt/src/new-ports/sysutils/sec/Makefile
> 15:USE_RC_SUBR=	sec
> 
> /opt/src/new-ports/sysutils/firstboot-growfs/Makefile
> 18:USE_RC_SUBR=	firstboot_growfs
> 
> /opt/src/new-ports/sysutils/syslog-ng319/Makefile
> 49:USE_RC_SUBR=	syslog-ng
> 
> /opt/src/new-ports/sysutils/rundeck/Makefile
> 22:USE_RC_SUBR=	rundeck
> 
> The reason for the option is to avoid the conflict. HEAD as of r354902 
> disables amd by default. The intention is to remove it by 13.0. There is a 
> subsequent (yesterday) update to param.h that would allow me to test 
> OSVERSION. Additionally people can still build and install world WITH_AMD, 
> the option gives us greater flexibility. This of course will change when 
> amd is eventually removed from HEAD.

Mmm, no, if you want to make it work correctly, do something like this:

.if !exists(/etc/rc.d/amd)
USE_RC_SUBR=	amd
.endif

Or use OSVERSION to determine if the file is needed or not, but an
options is really, really not the correct way to do it.

-- 
Mathieu Arnold
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 963 bytes
Desc: not available
URL: <http://lists.freebsd.org/pipermail/svn-ports-head/attachments/20191121/b1f2b244/attachment.sig>