x11/x11-libraries build

[Kris Kennaway]

On Tue, May 08, 2007 at 10:25:09PM -0400, Howard Goldstein wrote:
> Kris Kennaway wrote:
> >On Tue, May 08, 2007 at 07:33:49PM -0400, Howard Goldstein wrote:
> >>Kris Kennaway wrote:
> >>>On Tue, May 08, 2007 at 07:15:43PM -0400, Howard Goldstein wrote:
> >>>>Kris Kennaway wrote:
> >>>>>On Tue, May 08, 2007 at 06:58:47PM -0400, Howard Goldstein wrote:
> >>>>>>Heads up?  Hmmm.... make clean works, but not make:
> >>>>>>
> >>>>>>cally:/usr/ports/x11/xorg-libraries# make -v
> >>>>>>*** Error code 1
> >>>>>>
> >>>>>>Stop in /usr/ports/x11/xorg-libraries.
> >>>>>>cally:/usr/ports/x11/xorg-libraries#
> >>>>>>
> >>>>>>
> >>>>>>On a system with no /usr/X11R6 at all, something in the following 
> >>>>>>bits hoses the build when those vars are not defined:
> >>>>>>
> >>>>>>.if !defined(XORG_UPGRADE) && !defined(PACKAGE_BUILDING)
> >>>>>>pre-everything::
> >>>>>>      @test -d /usr/X11R6 && echo "Read ${PORTSDIR}/UPDATING for the 
> >>>>>>procedure to upgrade to xorg 7.2." && ${FALSE}
> >>>>>>.endif
> >>>>>Did you try reading the text you quoted?
> >>>>Yes.  Do you know why it's failing with that output, is it because 
> >>>>there is no /usr/X11R6? Keep in mind this is on a system that's been 
> >>>>tracking git for months.
> >>>Yes, see ${PORTSDIR}/UPDATING.
> >>Perhaps I'm not understanding your problem with my post.   Is the 
> >>argument to the echo supposed to display on this failure, or, is it 
> >>supposed to fail displaying "*** Error code 1" without any further 
> >>explanation which is what I'm seeing here?
> >
> >It is supposed to break if /usr/X11R6 is not a symlink.  So I guess it
> >should be
> >
> >@test -L /usr/X11R6 || ( echo "Read ${PORTSDIR}/UPDATING for the procedure 
> >to upgrade to xorg 7.2." && /usr/bin/false )
> 
> Yup, and no amount of my re-reading the UPDATING was going to fix that 
> logic :(
> 
> A slightly related issue hanging out there is the symlink vs the 
> local/etc/rc.d's.  It's not mentioned in the UPDATING and I didn't find 
> the Tools/mergebase.sh so I couldn't tell if something in there is 
> intended to patch rc.subr either with my nasty patch or some more artful 
> way of avoiding the problem with undesirable multiple local rc.d runs.

You're jumping the gun, it is yet to be committed.

Kris