svn commit: r258848 - user/hrs/releng/release

[Glen Barber]

On Tue, Dec 03, 2013 at 12:16:28AM +0900, Hiroki Sato wrote:
> gj> On Mon, Dec 02, 2013 at 02:40:26PM +0000, Hiroki Sato wrote:
> gj> > Author: hrs
> gj> > Date: Mon Dec  2 14:40:26 2013
> gj> > New Revision: 258848
> gj> > URL: http://svnweb.freebsd.org/changeset/base/258848
> gj> >
> gj> > Log:
> gj> >   Improve robustness and consistency, and remove redundant lines with no
> gj> >   major functional change:
> gj>
> gj> > [...]
> gj>
> gj> > +if [ $($ID -u) -ne 0 ]; then
> gj> > +	echo 1>&2 "$0: Needs to be run as root."
> gj> > +	exit 1
> gj> > +fi
> gj> >  # The directory within which the release will be built.
> gj> > -CHROOTDIR="/scratch"
> gj> > +: ${CHROOTDIR:=/scratch}
> gj> > +: ${CHROOT_CMD:=chroot $CHROOTDIR}
> gj> > +: ${DESTDIR=:/R/}
> gj> > +: ${MAKEOBJDIRPREFIX:=/usr/obj}
> gj> >
> gj>
> gj> Please do not merge these changes to head/.  It was intentional that
> gj> environment was not honored, as everything should be set in the
> gj> configuration file.
> gj>
> gj> It is possible to use the environment in the configuration file, such
> gj> as:
> gj>
> gj> ---- release.conf ----
> gj> : ${_DESTDIR=:/R/}
> gj> DESTDIR=${_DESTDIR}
> gj> ---- release.conf ----
> 
>  Everything can be overridden by a configuration file regardless of
>  whether envvars are honored even after this change.  What is the
>  problem?  I do not see negative effect assuming a conf file is used.
> 

This allows the host environment to accidentally pollute the build if
something is not set in configuration file and unintentionally set in
the builder's environment.

It was why I specifically did not allow the environment to affect the
build process.

Glen

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 834 bytes
Desc: not available
URL: <http://lists.freebsd.org/pipermail/svn-src-user/attachments/20131202/7bc906f0/attachment.sig>