[Top][All Lists]

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

Re: reorganized config.status

From: Ralf Wildenhues
Subject: Re: reorganized config.status
Date: Sat, 13 Aug 2005 00:00:13 +0200
User-agent: Mutt/1.5.9i

Hi Stepan,

* Stepan Kasal wrote on Wed, Aug 10, 2005 at 05:05:51PM CEST:
> On Wed, Aug 10, 2005 at 03:42:18PM +0200, Ralf Wildenhues wrote:
> > >   no, I don't think using ac_config_headers is a good idea.
> > 
> > I knew it.  ;)
> So I passed the test.  :-)

Other way round.  I tried to make you do the work for me and did not
succeed at first.  :)
(Also, I failed to see the first time that you only wanted me to do the
documentation and not the actual implementation..)

> > Rough draft, with a patch which does not fully work:
> Weird, I'd say that your patch is correct.
> Anyway, I'd hook it differently, see the attached patch.

I knew you would do that.  :)
Yours is much nicer, BTW.

> The documentation seems ok, but perhaps it is worth mentioning that
> AH_HEADER should be best used within code registered by

Hmm.  But this:
| m4_ifndef([AH_HEADER], [AC_CONFIG_HEADERS([config.h])])
| ])

would not be correct, right?
Quoting 'info Autoconf Configuration\ Headers:
| The `AC_CONFIG_HEADERS' macro selects this kind of output.  It should
| be called right after `AC_INIT'.

AC_CONFIG_HEADERS would be called way too late.
(Please remember that some people try to write their in a
way to be compatible with more than just the CVS version of Autoconf.)

But maybe I'm just overcautious here, and you have addressed all these
issues already.

> (Do I guess correctly that this is the case with libtool?)

Nope, ATM it isn't.  It's just called from within AC_LIB_LTDL aka
LTDL_INIT.  Guess we should change that, too.

> My biggest doubt is whether the name "AH_HEADER", which I coined without
> any thinking, is teh right one.  But no one has said anything, yet...
> Would you be so kind to combine the things together and test whether it
> works for you?

Works for me.  Below is the patch against Autoconf,
attached the one against Libtool (HEAD, branch-2-0).


        * lib/autoconf/status.m4 (_AC_CONFIG_REGISTER_DEST): New macro
        to define AH_HEADER.
        (_AC_CONFIG_FOOS): Use it.
        * doc/autoconf.texi (Configuration Headers): Document it.

Index: doc/autoconf.texi
RCS file: /cvsroot/autoconf/autoconf/doc/autoconf.texi,v
retrieving revision 1.925
diff -u -r1.925 autoconf.texi
--- doc/autoconf.texi   6 Jul 2005 21:39:31 -0000       1.925
+++ doc/autoconf.texi   12 Aug 2005 21:33:11 -0000
@@ -2697,6 +2697,18 @@
 to prepend and/or append boilerplate to the file.
 @end defmac
address@hidden AH_HEADER
+This macro is defined as the name of the first declared config header
+and undefined if no config headers have been declared up to this point.
+A third-party macro may, for example, require use of a config header
+without invoking AC_CONFIG_HEADERS twice like this:
+m4_ifndef([AH_HEADER], [AC_CONFIG_HEADERS([config.h])])
address@hidden example
address@hidden defmac
 @xref{Configuration Actions}, for more details on @var{header}.
Index: lib/autoconf/status.m4
RCS file: /cvsroot/autoconf/autoconf/lib/autoconf/status.m4,v
retrieving revision 1.72
diff -u -r1.72 status.m4
--- lib/autoconf/status.m4      26 Jul 2005 10:14:25 -0000      1.72
+++ lib/autoconf/status.m4      12 Aug 2005 21:33:11 -0000
@@ -258,9 +258,14 @@
 # With CONFIG_LINKS, reject DEST=., because it is makes it hard for 
 # to guess the links to establish (`./config.status .').
+# Save the name of the first config header to AH_HEADER.
 [_AC_CONFIG_UNIQUE([$1], [$3])dnl
-m4_if([$1 $3], [LINKS .], [AC_FATAL([invalid destination of a config link: 
+m4_if([$1 $3], [LINKS .],
+      [AC_FATAL([invalid destination of a config link: `.'])])dnl
+m4_if([$1], [HEADERS],
+      [m4_define_default([AH_HEADER], [$3])])dnl
 dnl Recognize TAG as an argument to config.status:

Attachment: ah-header-lt-HEAD
Description: Text document

reply via email to

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