[Rd] need help with libtool/aclocal error

[Prof Brian Ripley]

Looks like libtool is objecting to a patch-level mismatch in hte R 
sources (and I'm surprised that such minor mismatches are not 
allowed).  You should be able to solve this by using a current 
R-patched (>=r48719), or just updating tools/ltmain.sh from 
https://svn.r-project.org/R/branches/R-2-9-branch/tools/ltmain.sh and 
re-configuring and re-making R.

Looks like this has been mismatching for a while, so perhaps no one 
uses R CMD LINK.

On Fri, 5 Jun 2009, Kynn Jones wrote:

> I'm trying to build an executable for a program I wrote.  The
> compilation steps go well, but the linking step fails with a libtool
> "version mismatch" error.  My linking command has this prefix:
>
> /path/to/R/bin/R CMD LINK gcc -g -std=gnu99 ... etc.
>
> The error looks like this:
>
> libtool: Version mismatch error.  This is libtool 2.2, but the
> libtool: definition of this LT_INIT comes from libtool 2.2.6.
> libtool: You should recreate aclocal.m4 with macros from libtool 2.2
> libtool: and run autoconf again.
>
>
> I have no idea what this error message is referring to.  The version
> of libtool installed on the system is none of the ones mentioned in
> this error:
>
> % libtool --version
> ltmain.sh (GNU libtool) 1.5.22 Debian 1.5.22-4 (1.1220.2.365
> 2005/12/18 22:14:06)
>
> <FSF disclaimer>
>
>
>
> The libtool in the R directory has version 2.2:
>
> % /path/to/R/bin/libtool --version
>
> ltmain.sh (GNU libtool) 2.2
> Written by Gordon Matzigkeit <gord at gnu.ai.mit.edu>, 1996
>
> <FSF disclaimer>
>
>
> Strangely enough, this same libtool script has the following lines in
> its source code:
>
> # Which release of libtool.m4 was used?
> macro_version=2.2.6
> macro_revision=1.3012
>
>
>
> I don't know what to make of this.
>
> Be that as it may, does anyone know how I could get past this sticking point?
>
> TIA!
>
> G.
>
> P.S. I should point out that the R I'm using is one that I just
> compiled so that I could run R under valgrind.  Here's the ./configure
> command I used before running make:
>
> % ./configure --prefix=/path/to/R --enable-R-shlib --enable-BLAS-shlib
> --with-valgrind-instrumentation=2
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595