[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: AC_CYGWIN
From: |
Assar Westerlund |
Subject: |
Re: AC_CYGWIN |
Date: |
28 May 2003 20:47:40 -0400 |
User-agent: |
Gnus/5.09 (Gnus v5.9.0) Emacs/21.3 |
Patrick Welche <address@hidden> writes:
> I found the definition of AC_CYGWIN in lib/autoconf/specific.m4:
>
> # AC_CYGWIN
> # ---------
> # Check for Cygwin. This is a way to set the right value for
> # EXEEXT.
You will notice that this does not set EXEEXT.
> AU_DEFUN([AC_CYGWIN],
> [AC_CANONICAL_HOST
> AC_DIAGNOSE([obsolete],
> [$0 is obsolete: use AC_CANONICAL_HOST and $host_os])dnl
> case $host_os in
> *cygwin* ) CYGWIN=yes;;
> * ) CYGWIN=no;;
> esac
> ])# AC_CYGWIN
>
>
> Several things I don't understand: google brought up an email which
> esentially said "Don't use AC_REQUIRE inside AU_DEFUN", but then how
> do you say "Only run AC_CANONICAL_HOST if it hasn't been run before".
I think the point is that you shall not AC_REQUIRE inside an AU_DEFUN,
but you can AC_REQUIRE inside an AC_DEFUN.
Remember that an AU_DEFUN is expanded by autoupdate, so it should
really only contain the old macro written in terms of the new one.
> Also, AU_DEFUN (OLD-MACRO, IMPLEMENTATION, [MESSAGE]), so why use
> AC_DIAGNOSE instead of [MESSAGE]?
It seems that AC_CYGWIN has not been updated to take advantage of
3-argument AU_DEFUN.
Hope this helps.
- AC_CYGWIN, Patrick Welche, 2003/05/27
- Re: AC_CYGWIN,
Assar Westerlund <=