[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [RFC] Short- and medium-term proposal for AC_CHECK_* overhauling
From: |
Bonzini |
Subject: |
Re: [RFC] Short- and medium-term proposal for AC_CHECK_* overhauling |
Date: |
Thu, 1 May 2003 17:18:41 +0200 |
> > _ac_check_func 'strerror'
> > if test $ac_cv_func_strerror = yes; then
>
> Nice. How about this instead? it's even clearer:
>
> if ac_check_func strerror; then
I thought that this could give some portability headaches, plus the other
solution would fit better with the specialization concept (since in both
cases you end up with the cache variable).
> AC_USE_SHELL_FUNCTIONS(arg)
>
> If the argument is 1, use shell functions; if 0, don't use them. The
> default is 0 at first, but we'll change the default to 1 eventually.
I'll probably have to refine my m4 skills for this, but yes, it is possible of
course.
> > If parenthesized lists were ever to be made the only possibility,
>
> My own feeling is that it's too soon to require parenthesized lists.
> Users already have too many compatibility issues already.
That's why I wrote "were *ever* to be made the only possibility".
I don't think that's going to be done in less than... hmmm... two years.
Paolo
- Re: [RFC] Short- and medium-term proposal for AC_CHECK_* overhauling,
Bonzini <=