Re: pkg/38232

[Robert Elz <kre%munnari.OZ.AU@localhost>]

The following reply was made to PR pkg/38232; it has been noted by GNATS.

From: Robert Elz <kre%munnari.OZ.AU@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: pkg/38232 
Date: Sat, 22 Mar 2008 11:28:20 +0700

     Date:        Sat, 22 Mar 2008 01:20:03 +0000 (UTC)
     From:        Gavan Fantom <gavan%coolfactor.org@localhost>
     Message-ID:  <20080322012003.367BF63B863%narn.NetBSD.org@localhost>
 
   |  I'm not sure that pkg_comp should really need to perform the check if 
   |  pkgsrc is able to install pkg_install as a dependency.
 
 Deleting the check is certainly one possibility - I was assuming it was
 added for some rational reason, though perhaps with modern pkgsrc that
 reason no longer applies.
 
   |  Then, pkg_comp could set the variable when checking the pkg_install 
   |  versions, and be able to build pkg_install upfront until the bootstrap 
   |  dependency mechanism is more robust.
 
 It can (more or less) build it now I think - the problem is that it is
 deciding not to, when it probably should be building it.
 
 Then the other problem you described in your PR bites, and makes a mess
 of later compilations (which would almost certainly happen just the same
 way on a virgin 3.0 - it isn't pkg_comp that's causing that problem, pkg_comp
 just makes it easier to reproduce, over and over).
 
 kre