Re: test in a m4 macro and variable created from pushdef

[Vincent Torri]

On Fri, 10 Aug 2007, Ralf Wildenhues wrote:

> Hello Vincent,
>
> * Vincent Torri wrote on Fri, Aug 10, 2007 at 07:51:00AM CEST:
> > On Thu, 9 Aug 2007, Eric Blake wrote:
> >
> > >  AM_CONDITIONAL([BUILD_DRIVER_]UP, [...])
> > > calls AM_CONDITIONAL with a string consisting of UP already expanded (ok
> > > if UP is alphanumeric, but with the potential for m4 syntax confusion if
> > > it contains anything in the character class [][,()].
> >
> > both solutions:
> >
> > AM_CONDITIONAL([BUILD_DRIVER_]UP, [test "x${use_]DOWN[_driver}" = "xyes"])
> > AM_CONDITIONAL([BUILD_DRIVER_]UP, test "x${[use_]DOWN[_driver]}" = "xyes")
> >
> > are working
>
> One check to avoid further hidden issues is to
>  grep UP configure
>  grep DOWN configure
>
> There seems to be at least one harmless unexpanded instance left in a
> comment, hopefully fixed by the (untested) patch below.  OK to apply?

Finally, I've compile autoconf (cvs). Remark that the documentation fails 
on mingw. If you want the error, I can paste it somewhere.

I've applied the patch, installed autoconf, launched autoreconf, used both 
syntax:

AM_CONDITIONAL([BUILD_DRIVER_]UP, [test "x${use_]DOWN[_driver}" = "xyes"])
AM_CONDITIONAL([BUILD_DRIVER_]UP, test "x${[use_]DOWN[_driver]}" = "xyes")

but nothing specific occurs.

what should that patch do (i'm not an expert at all of m4) ?

Vincent Torri