[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Testing for a GUILE installation
From: |
David Allouche |
Subject: |
Testing for a GUILE installation |
Date: |
Thu, 10 Oct 2002 15:02:32 +0200 |
User-agent: |
Mutt/1.4i |
On Thu, Oct 10, 2002 at 02:32:31PM +0200, David Allouche wrote:
>
> I am currently working on porting TeXmacs to work with guile-1.6.0.
>
> I happened to notice that when you use "guile-config info top_srcdir"
> you get the string "top_srcdir_absolute@" instead of something like
> "/home/david/home/devel/guile-1.6.0".
But, you may ask, what is the use of top_srcdir to TeXmacs? Well,
TeXmacs goes to some length to test that GUILE is properly installed.
What it does is test for the existence of "ice-9/boot-9.scm" in
GUILE_LOAD_PATH. If that fails, it looks for a default GUILE_LOAD_PATH
with the following procedure:
datadir=`guile-config info datadir`
version=`guile-config info top_srcdir`
Drop substring of version up to the last '-'
Drop substring of version up to the first '/'
(Here $version is supposedly the complete version number)
Prepend "$datadir/guile/$version" to GUILE_LOAD_PATH
Look for "ice-9/boot-9.scm" in GUILE_LOAD_PATH.
I worked around the current bug in guile-config, by using
'guile --version' and doing a different parsing of the result. But I
would like to know if there is a recommended, less hackish, way of
testing that GUILE is correctly installed.
Also, for TeXmacs to work with guile-1.6.0 I have to avoid prepending
the "$datadir/guile/$version" to GUILE_LOAD_PATH, otherwise I get an
error:
ERROR: Unbound variable: include-deprecated-features
Leaving GUILE_LOAD_PATH alone seems to work okay also with guile-1.4.
But then I wonder what was the use of that code...
FYI, I configured guile with --enable-deprecated=detailed.
--
David Allouche | GNU TeXmacs -- Writing is a pleasure
Free software engineer | http://www.texmacs.org
http://ddaa.net | http://alqua.com/tmresources
address@hidden | address@hidden
TeXmacs is NOT a LaTeX front-end and is unrelated to emacs.