[R] readline with R-1.4.0

[Roger Peng]

This was from a previous post to this list:

-- CLIP ---

This was discussed a few days ago on this very list. The correct
answer for redhat 7.2 is to install

readline41-4.1-10.i386.rpm

By definition, symbolic libraries with different "sonames" (like
/usr/lib/libreadline.so.4.1 and ...4.2) may be incompatible, so
programs compiled for one of them shouldn't try to use the other. (Some
packages insist on letting the soname follow the release name whether
or not versions are binary compatible so there's no guarantee it would
break either.)

Empirically it seems to work for R if you make the .4.1 lib a symlink
to the .4.2 lib, but there is no telling what that might do to other
programs.

Trying "rpm -i --force ...." will most likely crash R on startup.

-- CLIP --

-roger
_______________________________
UCLA Department of Statistics
rpeng at stat.ucla.edu
http://www.stat.ucla.edu/~rpeng

On Wed, 19 Dec 2001, apjaworski at mmm.com wrote:

> Hi,
> 
> I am not sure if this is a bug, so I am posting it here.  I also searched
> the FAQ file for the readline string and did not find anything about the
> behavior below.
> 
> I just successfully rsync'ed, compiled and tested R-1.4.0 on my RH 7.2
> Linux machine.  I encountered one problem though.  When running ./configure
> I noticed that I was getting errors from the readline library.  Although
> configure detected readline, the internal test it runs failed.  Now, my RH
> distribution has readline version 4.2-2 (both dynamic and static).  It also
> contains the older version for compatibility purposes.  In /usr/lib I have
> 
>      libreadline.a                       (static 4.2-2)
>      libreadline.so.4.2            (dynamic 4.2-2)
>      libreadline.so.3.0            (dynamic 2.2.1 !?)
>      libreadline.so -> libreadline.so.4.2     (symbolic link)
> 
> This setup created the failure I just mentioned.  However, when I changed
> the link to
> 
>      libreadline.so -> libreadline.so.3.0
> 
> everything went fine with configure and compilation.  The odd thing is that
> after compilation finished I changed the above symbolic link back to
> libreadline.so.4.2 and all tests ran with no problem.  Readline also seems
> to work in interactive mode.
> 
> Andy
> 
> __________________________________
> Andy Jaworski
> Engineering Systems Technology Center
> 3M Center, 518-1-01
> St. Paul, MN 55144-1000
> -----
> E-mail: apjaworski at mmm.com
> Tel:  (651) 733-6092
> Fax:  (651) 736-3122
> 
> -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
> r-help 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-help-request at stat.math.ethz.ch
> _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
> 

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help 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-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._