libtool
[Top][All Lists]
Advanced

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

Re: getting libtool to not call AC_PROG_FC and AC_PROG_F77...


From: Ralf Wildenhues
Subject: Re: getting libtool to not call AC_PROG_FC and AC_PROG_F77...
Date: Thu, 18 Jan 2007 13:41:29 +0100
User-agent: Mutt/1.5.13 (2007-01-09)

* Mike Frysinger wrote on Tue, Jan 16, 2007 at 11:48:31PM CET:
> On Tuesday 16 January 2007 17:30, Ed Hartnett wrote:
> > When I call AC_PROG_LIBTOOL it seems to call AC_PROG_F77 and
> > AC_PROG_FC, but I don't want it to.
> 
> this should be fixed in the latest cvs, just not latest release

Bzzt.  Wrong.  (Also note that Ed is using a CVS version that's new
enough.)

> > Is there any way to tell libtool that I've already called these
> > macros, and they should not be called again?
> 
> searching the mailing list leads me to:
> http://lists.gnu.org/archive/html/libtool/2005-09/msg00098.html

Yes, but that's the case where someone has no F77 at all, and wants no
F77 tests at all, rather than Ed's situation: he has and wants F77 and
FC, just not in the order that the Libtool macros dictate.

The simplest answer I know of for the case where one has no F77/CXX at
all (besides backporting the HEAD fix for the issue) is
  m4_defun([_LT_AC_LANG_CXX_CONFIG])
  m4_defun([_LT_AC_LANG_F77_CONFIG])

which happens to magically not break when you move to Libtool 2.0
eventually.  Again, this does not apply to Ed who already uses CVS HEAD.

Incidentally, the code in the link you posted could help Ed.  But I'll
add that there is no way we can guarantee this to remain functional.
I don't know if it's feasible for Libtool to guarantee that you can
split the compiler and its shared library creation bits.  Someone with
time could try to look into it...

Cheers,
Ralf




reply via email to

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