autoconf-patches
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 16-ac-lang-preproc-require.patch


From: Pavel Roskin
Subject: Re: 16-ac-lang-preproc-require.patch
Date: Sun, 12 Nov 2000 13:39:00 -0500 (EST)

Hello, Akim!

> Index: ChangeLog
> from  Akim Demaille  <address@hidden>
>       Set AC_LANG_PREPROC_REQUIRE which replaces AC_REQUIRE_CPP.
>       The main difference is that the former requires AC_LANG_COMPILER.

Ok.

>       * aclang (AC_LANG_PREPROC, AC_LANG_PREPROC(C), AC_LANG_PREPROC(C++))

Ok.

>       (AC_LANG_PREPROC(Fortran 77), AC_LANG_PREPROC_REQUIRE): New
>       macros.  Issue a warning when looking for the Fortran 77
>       preprocessor instead of an error.

I'm not sure that we really want to go ahead after that warning. Howvalid
"configure" will be if we ignore it?

>       (AC_LANG_COMPILER): Check that it is run before the corresponding
>       AC_LANG_PREPROC.
>       (AC_PROG_C
>       (AC_LANG_COMPILER_REQUIRE): Don't call directly
>       AC_LANG_COMPILER(_AC_LANG), rather invoke AC_LANG_COMPILER so that
>       the generic code in AC_LANG_COMPILER is run.
>       (AC_REQUIRE_CPP): Use AC_LANG_PREPROC_REQUIRE.
>       (AC_PROG_CC, AC_PROG_CXX): Don't require being run before the
>       corresponding AC_LANG_PREPROC: AC_LANG_COMPILER does it.

Ok.

>       Propagate AC_LANG_PREPROC_REQUIRE.
> 
>       * acgeneral.m4 (AC_TRY_CPP, AC_EGREP_CPP): Use it instead of
>       AC_REQUIRE_CPP.
>       * acspecific.m4 (_AC_DECL_YYTEXT, AC_PATH_X): Don't
>       AC_REQUIRE_CPP, inner macro will do it.
>       * aclang.m4 (AC_PROG_GCC_TRADITIONAL, AC_C_STRINGIZE)
>       (AC_C_PROTOTYPES): Likewise.

Ok.

>       (AC_C_STRINGIZE): Yeeks!  The body of AC_CACHE_CHECK was not
>       quoted.  Use @%:@ do assist Emacs.

I recently went through all occurences of `))' but missed this somehow.
Shame on me.

>       For some reason (don't ask), this revamping revealed that
>       AC_PROG_CC_STDC, because of the `break', does not clean its tmp
>       files.

Well, we don't test the macros that are AC_REQUIRE'd. This doesn't seem to
be a very clean appoach, since the caller can cleanup for the callee.
mktests.sh may need to be changed.

Anyway, make sure to regenerate syntax.at and commit it.

>       * aclang.m4 (AC_PROG_CC_STDC): Extract the creation of conftest.c
>       out of AC_COMPILE_IFELSE.
>       Be sure to clean the tmp files.

Ok.

> Index: aclang.m4
> --- aclang.m4 Sat, 11 Nov 2000 18:31:28 +0100 akim (ace/b/32_aclang.m4 
> 1.5.3.47 644)
> +++ aclang.m4 Sun, 12 Nov 2000 12:33:44 +0100 akim (ace/b/32_aclang.m4 
> 1.5.3.47 644)
> @@ -1,6 +1,6 @@
>  # This file is part of Autoconf.                       -*- Autoconf -*-
>  # Programming languages support.
> -# Copyright 1992, 93, 94, 95, 96, 98, 99, 2000
> +# Copyright 2000

You meant 1992-2000?

> +# AC_REQUIRE_CPP
> +# --------------
> +# Require the preprocessor for the current language.
> +# FIXME: AU_ALIAS once AC_LANG is officially documented (2.51?).

Huh? AC_LANG is documented.

> +# AC_LANG_PREPROC(Fortran 77)
> +# ---------------------------
> +# Find the C compiler.  Must be AC_DEFUN'd to be AC_REQUIRE'able.

Find the C compiler? Make sure to change the comment.

> +AC_DEFUN([AC_LANG_PREPROC(Fortran 77)],
> +[m4_warn([syntax],
> +         [$0: No preprocessor defined for ]_AC_LANG)])

See above. This should be a fatal error unless I'm missing something.

Everything else seems to be Ok.

Regards,
Pavel Roskin





reply via email to

[Prev in Thread] Current Thread [Next in Thread]