Re: Problem with AC_REQUIRE expansion order

[Eric Blake]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

[dropping bug-autoconf]

According to Ralf Wildenhues on 8/5/2008 1:43 PM:
|
|>  AC_DEFUN([AC_PROG_CC],
|> +[AC_REQUIRE([_AC_PROG_CC])dnl
|> +m4_expand_once([m4_divert_text([DEFAULTS],
|> +  [ac_prog_cc_list=])])dnl
|> +m4_ifval([$1], [m4_divert_text([INIT_PREPARE],
|> +  [ac_prog_cc_list='$1'])])dnl
|
| This won't like
|   AC_PROG_CC(['cc -m32' "cc -m64"])

Ouch.  And that looks like a real possibility of being in existing code.
Oh well, time to update the testsuite to exercise this usage.

|
| and due to how AC_CHECK_TOOLS works, it's not fixed by a simple
| bpatsubst either.  Sorry.

And in light of the suggestion in this thread [1], I also have to support
existing code that does:
AC_PROG_CC([$mylist])

[1] http://lists.gnu.org/archive/html/bug-autoconf/2007-03/msg00012.html

Sounds like an eval is in order, along with a m4_bpatsubsts([[$1]], ['],
['\'']) when assigning ac_prog_cc_list, but where exactly to stick the eval?

eval 'for prog in '"$ac_prog_cc_list"'; do echo "$prog" ; done'

works in isolation, but when it comes to AC_CHECK_TOOLS, there's really no
easy way to surround the entire macro inside an eval.  So I guess I'm
stuck with:

eval set x "$ac_prog_cc_list"
shift
AC_CHECK_TOOLS(["address@hidden"])

I hope AC_PROG_CC is called after the point at which we can't nuke 
address@hidden
Meanwhile, I'll try to build up a test case using your example.

- --
Don't work too hard, make some time for fun as well!

Eric Blake             address@hidden
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (Cygwin)
Comment: Public key at home.comcast.net/~ericblake/eblake.gpg
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkiZE8EACgkQ84KuGfSFAYD+qgCfYaIZ1NrXt0ZrqlLdJdxYWhrz
Du0AnAxr1EoGAxwQ84XMp3ZirTWebytT
=SraX
-----END PGP SIGNATURE-----