Re: [Rd] Cross compiling in R-2.0.0

From: Prof Brian Ripley <ripley_at_stats.ox.ac.uk>
Date: Tue 19 Oct 2004 - 21:12:01 EST


On Tue, 19 Oct 2004, Iago Mosqueira wrote:

> When compiling R-2.0.0 for cross-compiling,

More likely you are cross-compiling R to run under Windows.

> I get the following error when calling 'make R'

Who said you should `make R'? That is not in the instructions in src/gnuwin32/INSTALL. If you do that you should get

gannet% make R
make: *** No rule to make target `R'. Stop.

> ---------- Making package base ------------
> adding build stamp to DESCRIPTION
> /bin/sh: line 1: --vanilla: command not found
> make[5]: *** [frontmatter] Erro 127
> make[4]: *** [all] Erro 2
> make[3]: *** [pkg-base] Erro 2
> make[2]: *** [rpackage] Erro 2
> make[1]: *** [all] Erro 2
> make[1]: Saíndo do directorio `/home/iago/crossbuild/WinR/R-2.0.0/src/gnuwin32'
>
> The error appears in the call to $(REXE), that was defined in
> R-2.0.0/src/gnuwin32/MakePkg (Line 18) as:
>
> REXE = $(R_EXE) --vanilla --slave
>
> I tried substituting REXE with R --vanilla --slave and that did the trick, only
> to later fail with a complain about $(R_HOME).

So R_EXE is not defined. Setting it is part of the instructions.

> Is there any reason for this failure in the Makefile variables? Anything to do
> with the change from 1.9.1 to 2.0.0?

User error! If you follow the instructions it does work, at least in an English-speaking locale.

-- 
Brian D. Ripley,                  ripley@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

______________________________________________
R-devel@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Received on Tue Oct 19 21:22:49 2004

This archive was generated by hypermail 2.1.8 : Fri 18 Mar 2005 - 09:00:37 EST