Skip site navigation (1)Skip section navigation (2)
Date:      31 Jul 1997 09:39:40 +0200
From:      Thomas Gellekum <tg@ihf.rwth-aachen.de>
To:        "Jordan K. Hubbard" <jkh@time.cdrom.com>
Cc:        scott@statsci.com, freebsd-ports@FreeBSD.ORG
Subject:   Re: how to handle compiled in paths?
Message-ID:  <87204f8ybn.fsf@ghpc6.ihf.rwth-aachen.de>
In-Reply-To: "Jordan K. Hubbard"'s message of Thu, 31 Jul 1997 00:19:14 -0700
References:  <20201.870333554@time.cdrom.com>

next in thread | previous in thread | raw e-mail | index | archive | help
"Jordan K. Hubbard" <jkh@time.cdrom.com> writes:

> > I was just noticing something with my nmh port...it seems that the prefix is 
> > compiled into the binaries, so if you make a package out of it, you can't
> > retarget the prefix with the pkg_add -p option (without recompilation, which
> > kind of defeats the purpose of the package & its -p option).  How is this
> > kind of stuff normally handled?
> 
> You repeat the @cwd line twice at the top of the package's PLIST,
> "stapling" it to the canned prefix and preventing a -p override.  Not
> every user's desire, of course, this still at least ensures that
> what's installed will work and is probably the best that can be done
> under the circumstances.

I think the `-p' option should just go away. We have hundreds of ports
with compiled in paths, most of which are dealt with at compile
time. Splitting the ports up into those that can be moved and those
that have to be recompiled is more of a pain than simply saying ``The
provided packages install into /usr/local, period. Please recompile if
you want another directory prefix.''.

tg



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?87204f8ybn.fsf>