Re: [Maxima-commits] CVS: maxima configure.in,1.35,1.36
Subject: Re: [Maxima-commits] CVS: maxima configure.in,1.35,1.36
From: James Amundson
Date: Mon, 24 Nov 2003 21:38:48 -0600
On Mon, 2003-11-24 at 11:54, Raymond Toy wrote:
> >>>>> "James" == James Amundson <amundson@users.sourceforge.net> writes:
>
> James> [CMUCL_NAME=${cmucl_default_name}])
> James> +AC_ARG_WITH(cmucl-runtime,
> James> + [ --with-cmucl-runtime=<path> Use cmucl runtime <path> (default
> James> + *cmucl-lib*/../bin/lisp)],
> James> + [cmucl=true
> James> + CMUCL_RUNTIME_PATH="$withval"
> James> + CMUCL_RUNTIME=`basename ${CMUCL_RUNTIME_PATH}`],
> James> + [CMUCL_RUNTIME_PATH=`${CMUCL_NAME} -quiet -batch -eval '(format t "~a/../bin/lisp~%" common-lisp::*cmucl-lib*)(quit)'`
>
> You might want to add -noinit and -nositeinit. Not sure.
I've wondered about that. I think we should add them. (Done.)
> But anyway, this doesn't work in current CVS versions because
> *cmucl-lib* can be a list of directories, like $PATH.
OK. I'm not surprised. I was uncomfortable adding something as obviously
implementation-dependent as the above, but I don't see any way around it
if we want to support static installs. (There is a similar issue with
clisp.) I think static installs will be a big win in the long run, so
the additional complications should be worth it.
The automatic runtime determination is just a guess -- it can always be
overridden by an explicit choice.
> I'll try to think of some way to get this information out.
Thanks. I'd welcome any suggestions.
--Jim