Re: LIBOBJS and ANSI2KNR fix?

[Harlan Stenn]

> >>>>> "Harlan" == Harlan Stenn <address@hidden> writes:
> 
> Harlan> Thanks for your response.
> >> > From: Harlan Stenn <address@hidden> > Date: Sun, 24 Mar
> >> 2002 16:54:14 -0500
> >> > 
> >> > - Do I still need to worry about this problem with LIBOBJS and
> >> ANSI2KNR?
> >> 
> >> I think the answer is "no".
> 
> Harlan> I did a quick test tonight, and it's fixed in 2.52.  I'm not
> Harlan> sure about 2.50.
> 
> Err, I don't know what you name `fixed'.

I (thought I) tested it and it worked.  I got your email, tested it again,
and saw that it *didn't* work!

Sigh...

> `AC_LIBOBJ' vs. `LIBOBJS'
> -------------------------
> 
> ...  Because the token `LIBOBJS' is now forbidden,
> you will have to replace this snippet with:
> 
>      # This is necessary so that .o files in LIBOBJS are also built via
>      # the ANSI2KNR-filtering rules.
>      LIB@&address@hidden "$LIB@&address@hidden" |
>                   sed 's,\.[[^.]]* ,$U&,g;s,\.[[^.]]*$,$U&,'`
>      LTLIBOBJS=`echo "$LIB@&address@hidden" |
>                 sed 's,\.[[^.]]* ,.lo ,g;s,\.[[^.]]*$,.lo,'`
>      AC_SUBST(LTLIBOBJS)

I was concerned that this hack would lose under 2.53 because the addition of
$U might happen twice.

I've got a nasty throat infection and head cold so I couldn't test this
theory of mine; I solved the problem by not using LIBOBJS; I now #ifdef out
the guts of the subroutines I was previously handling via
AC_REPLACE_FUNCS().

H