[Rd] configure problem in R-devel caused by conditionals

Simon Urbanek simon.urbanek at math.uni-augsburg.de
Sun Sep 14 19:29:53 MEST 2003

I re-generated configure of the latest R-devel with autoconf 2.57 and 
got the following problem:

configure: error: conditional "HAVE_ORBIT" was never defined.
Usually this means the macro was only invoked conditionally.

(resolving HAVE_ORBIT leaves us with HAVE_GNORBA causing the same 
problem. You should be able to reproduce it by running autoreconf (or 
aclocal; autoconf - not just autoconf!) and configure on the r-devel 

The culprit is primarily located in acinclude.m4, namely the macro 
afaik it's never used in configure.ac (I couldn't find it at least), 
but the conditionals are "seen" by autoconf. Since all conditionals 
must be properly defined, it bails out.
(from automake docs: The shell condition (suitable for use in a shell 
if statement) is evaluated when configure is run.  Note that you must 
arrange for every AM_CONDITIONAL to be invoked *every* time configure 
is run - if AM_CONDITIONAL is run conditionally (e.g., in a shell if 
statement), then the result will confuse automake. )

Removing those macros from acinclude.m4 resolves the issue.

But, I'm wondering: in the m4 directory the README states that the 
files in m4 are used to create acinclude.m4 - is that done by hand or 
is there any "official" way to do it? There is no script called 
'bootstrap' in the root dir so the whole autoconf process seems to be 
non-standartized ... (ok, in the develper docs it says that one should 
simply run autoconf .. but then such issues as the above won't become 
visible as autoconf doesn't call aclocal, whereas e.g. autoreconf does 
...). So, how is the 'official' configure created?

Simon Urbanek
Department of computer oriented statistics and data analysis
Universitätsstr. 14
86135 Augsburg

Tel: +49-821-598-2236
Fax: +49-821-598-2280

Simon.Urbanek at Math.Uni-Augsburg.de

More information about the R-devel mailing list