R devel - Jun 2000 - Missing -lm for tcltk (R 1.1.0, AIX 4.3) (PR#573)

> From: tov@phoenix.ece.cmu.edu
> Date: Fri, 16 Jun 2000 16:32:36 +0200 (MET DST)
> To: r-devel@stat.math.ethz.ch
> Subject: [Rd] Missing -lm for tcltk (R 1.1.0, AIX 4.3) (PR#573)
> CC: R-bugs@biostat.ku.dk
> X-Loop: R-bugs@biostat.ku.dk
> 
> Hi,
> 
>   ok, AIX 4.3 insists on being different.  I can't compile --with-tcltk
>   out of the box.  I have to add '-lm' to the SHLIBLDFLAGS manually
>   after configure.
Yes, I guess that's because it insists on resolving libm at the shared
libs build stage.
>   Which brings up the question:  Can you use TK_LIBS instead of TK_XLIBSW
>   during configure? Excerpt from tkConfig.sh on my machine:
> 
> # Linker switch(es) to use to link with the X11 library archive.
> TK_XLIBSW='-lX11' 
> 
> # Additional libraries to use when linking Tk.
> TK_LIBS='-lX11 -lld  -lm' 
> 
>   The '-lld' wouldn't hurt, would it?
>   Maybe the following would work for configure?
> 
> case "${host}" in
>   *aix*)
>     LIBS="-ltk${TK_VERSION} ${TCLTK_LIBS} ${TK_LIBS}  $LIBS"
>     ;;      
>   *)
>     LIBS="-ltk${TK_VERSION} ${TCLTK_LIBS} ${TK_XLIBSW}  $LIBS"
>     ;;      
> esac      
It's not done like that. Take a look in aclocal.m4.  I suspect that
${TK_LIBS} would do everywhere, but will need to do some testing.

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

in the windows version
> z<-1:10
> dataentry(z)
Error in dataentry(z) : Argument "modes" is missing, with no default
according to the docs,
Modes=NULL should be default



--
Erich Neuwirth, Computer Supported Didactics Working Group
Visit our SunSITE at http://sunsite.univie.ac.at
Phone: +43-1-4277-38624 Fax: +43-1-4277-9386
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Prof Brian Ripley <ripley@stats.ox.ac.uk> writes:
> > From: tov@phoenix.ece.cmu.edu
> >   ok, AIX 4.3 insists on being different.  I can't compile
--with-tcltk
> >   out of the box.  I have to add '-lm' to the SHLIBLDFLAGS
manually
> >   after configure.
> 
> Yes, I guess that's because it insists on resolving libm at the shared
> libs build stage.
And the symbols from libm don't make it into R.exp.
> >   Which brings up the question:  Can you use TK_LIBS instead of
TK_XLIBSW
> >   during configure? Excerpt from tkConfig.sh on my machine:
> > 
> > # Linker switch(es) to use to link with the X11 library archive.
> > TK_XLIBSW='-lX11' 
> > 
> > # Additional libraries to use when linking Tk.
> > TK_LIBS='-lX11 -lld  -lm' 
> > 
> >   The '-lld' wouldn't hurt, would it?
> >   Maybe the following would work for configure?
> > 
> > case "${host}" in
> >   *aix*)
> >     LIBS="-ltk${TK_VERSION} ${TCLTK_LIBS} ${TK_LIBS}  $LIBS"
> >     ;;      
> >   *)
> >     LIBS="-ltk${TK_VERSION} ${TCLTK_LIBS} ${TK_XLIBSW} 
$LIBS"
> >     ;;      
> > esac      
> 
> It's not done like that. Take a look in aclocal.m4.  I suspect that
> ${TK_LIBS} would do everywhere, but will need to do some testing.
Ok, aclocal.m4 was obviously new to me.  Below is the patch.  It
works for me...  

With this patch and "SHLIBLD=CC ./configure" I can compile, install,
check R 1.1.

Regards,
  -tom


*** aclocal.m4  Mon Jun 19 11:39:56 2000
--- aclocal.m4.orig     Mon Jun 19 11:34:21 2000
***************
*** 840,847 ****
            if test -n "${TK_CONFIG}"; then
              . ${TK_CONFIG}    # get TK_VERSION
              AC_CHECK_LIB(tk${TK_VERSION}, Tk_Init,
!               [ TCLTK_LIBS="${TCLTK_LIBS} -ltk${TK_VERSION} 
${TK_LIBS}"
!                 have_tcltk=yes ], , [${TCLTK_LIBS} ${TK_LIBS}] )
            fi
          fi
        fi
--- 840,847 ----
            if test -n "${TK_CONFIG}"; then
              . ${TK_CONFIG}    # get TK_VERSION
              AC_CHECK_LIB(tk${TK_VERSION}, Tk_Init,
!               [ TCLTK_LIBS="${TCLTK_LIBS} -ltk${TK_VERSION} 
${TK_XLIBSW}"
!                 have_tcltk=yes ], , [${TCLTK_LIBS} ${TK_XLIBSW}] )
            fi
          fi
        fi

-- 
mailto:tov@ece.cmu.edu (Tom Vogels)   Tel: (412) 268-6638   FAX: -3204
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._