Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 20 Nov 1996 11:10:57 +0100 (MET)
From:      Martin Cracauer <cracauer@cons.org>
To:        freebsd-ports@freebsd.org
Subject:   And another question
Message-ID:  <199611201010.LAA02255@knight.cons.org>

next in thread | raw e-mail | index | archive | help

In Common Lisp, when installing a library as an addition to an
existing implementation, the compiled files are often saved into a new
Lisp world. Usually, this new world is the default so that it is not
possible to run the implementation without the additional stuff
anymore. 

I don't like this and usually install seperate executables for running
the implementation with or without additional stuff.

FreeBSD maintainers, do you have any preferences how I should handle
this in ports?

I think it is right to have additional executables, but what is the
mechanism to tell the user about it (if any)? Example problem: User
installed the base implemenation and one additional library for - say
- Xlib access. Now he goes and want to install another package.

The user must decide whether he/she wants the second additional library
to be build into the plain image and/or into the image already
containing the first additional library.

Then, how do I handle this in a port? Should I search for installed
implementations, then for alternate saving of it containing other
libraries and for every item ask the user whether to extend that one?
I'd rather like to have a mechanism that is non-interactive and puts
up everything overnight.

If I install several executables, how should I name them? I think of
the following, please drop me a flame if you don't link it.
- clisp                is the basic executable
- clisp-clx            the same with Xlib loaded
- clisp-clx-garnet     the same with Xlib and garnet loaded

Then, what do I do with basic implementations that already have an
extension installed? To follow this naming scheme, there should be a
'clisp-clos', but 'clisp' already contains 'clos'. Should I make a
symbolic or hard link to keep the naming scheme or should I trust the
user to look up the port description to find out what the basic system
provides?

The whole thing is quite challenging, but it helps to make my own mind
up :-)

Martin
-- 
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Martin Cracauer <cracauer@cons.org> http://www.cons.org/cracauer
  cracauer@wavehh.hanse.de (batched, preferred for large mails)
  Tel.: (daytime) +4940 41478712 (sometimes hacker's daytime :-)
  Tel.: (private) +4940 5221829 Fax.: (private) +4940 5228536
  Paper: (private) Waldstrasse 200, 22846 Norderstedt, Germany



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199611201010.LAA02255>