[1686] in Kerberos_V5_Development
Re: something useful
daemon@ATHENA.MIT.EDU (Theodore Y. Ts'o)
Sat Aug 31 00:48:57 1996
Date: Sat, 31 Aug 1996 00:48:20 -0400
From: "Theodore Y. Ts'o" <tytso@MIT.EDU>
To: Marc Horowitz <marc@MIT.EDU>
Cc: "Theodore Y. Ts'o" <tytso@MIT.EDU>, Ken Raeburn
<raeburn@cygnus.com>,
Kev <klmitch@MIT.EDU>, krbdev@MIT.EDU
In-Reply-To: Marc Horowitz's message of Fri, 30 Aug 1996 19:08:44 EDT,
<199608302308.TAA00662@beeblebrox.MIT.EDU>
Cc: Ken Raeburn <raeburn@cygnus.com>, Kev <klmitch@MIT.EDU>, krbdev@MIT.EDU
Date: Fri, 30 Aug 1996 19:08:44 EDT
I don't think you need to go as far as calling another Makefile. You
could have a platform-dependent switch in aclocal.m4, called from
configure.in, which set a few variables...
Yes, there are some gross things which might or might not work; I'm not
convinced that it's possible to accomodate *all* of different types of
shared libraries, by using @FOO@ substitutions and makefile substitution
rules, but I'll reserve judgement on that score.
However, I'll note with amusement that it wasn't all that long ago that
you and Barry were swearing up and down that making the entire Krb5 tree
depend on gmake was the only way to make the build system and the
current Makefile scheme "sane", and that the main feature which you
needed from the gmake system was the ability to expand macros.
Now, when we come up with a scheme which would allow autoconf to expand
simple macros, now you start arguing that perhaps autoconf + standard
make might be enough after all.
Then again, maybe your long deserved vacation made you see the light. :-)
- Ted
