Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 17 Aug 1998 15:16:54 -0700 (PDT)
From:      asami@FreeBSD.org (Satoshi Asami)
To:        ac199@hwcn.org
Cc:        obrien@NUXI.com, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org
Subject:   Re: cvs commit: ports/lang/glibstdc++28 Makefile
Message-ID:  <199808172216.PAA25476@silvia.hip.berkeley.edu>
In-Reply-To: <Pine.BSF.3.96.980817172226.1050F-100000@localhost> (message from Tim Vanderhoek on Mon, 17 Aug 1998 17:47:19 -0400 (EDT))

next in thread | previous in thread | raw e-mail | index | archive | help
 * Well, from a simply pragmatic perspective, this gives ports a
 * chance twiddle their final plist.  Or, perhaps a chance to read
 * their final plist potentially as an installation guide.

That's what I thought too, but David's probably right in that the
plist generation should be done right before it's used.

 * Are their any ports that legitimately need to modify their plist
 * during *-install?  Usually installation decisions are made during
 * *-configure...

For some ports, it might be easier to just install everything and run
a find.  This means there probably should be at least one
porter-definable install target before the plist is generated by
bsd.port.mk.  (It can take the porter-generated plist as an input.)

Where should I put it?  Right before fake-pkg (after post-install)
won't leave any chance for the port to do anything with it.  Maybe
between do-install and post-install?

Satoshi



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199808172216.PAA25476>