[R] perl advice

John D. Barnett jbarnett at wi.mit.edu
Tue Dec 7 21:00:50 CET 1999


Thanks; this is all very helpful.

I looked into Comm.pl, but had problems using it.  I found a message somewhere
saying that Comm.pl was fairly specific to SunOS 4/5; there's a newer perl
module
with similar functionality, Expect.pm

I've already written a routine to parse R output produced by dump(), but now
that
I'm able to interact directly, I'd like to parse R's normal output-- unless
there's a
way that I can send dump's output to STDOUT.

Thanks again!

Prof Brian D Ripley wrote:

> On Mon, 6 Dec 1999, Guido Masarotto wrote:
>
> > On Mon, Dec 06, 1999 at 05:12:44PM -0500, John D. Barnett wrote:
> > > Hello-
> > >
> > > Does anyone have a recommendation on how to call R from perl?  I'm using
> > > the IPC::Open2 module, and running R with the --slave and --quiet
> > > options.  The problem is that I can't predict how many lines of output I
> > > should try to read for each command-- if any!
> > >
> > > The ultimate goal is to use perl to provide a form-driven web interface,
> > > but have R do the underlying calculations.
> > >
> >
> >   A trick that I have used sometimes from tcl/tk is to redefine the
> >   prompt to some funny string (sending to the controlled process
> >   a 'option(prompt=...) command at the beginning) and then wait
> >   for the prompt on the output. If I remember also ESS
> >   works in a  probably less weaker but similar approach (i.e., it
> >   waits for a particular regular expression on the standard output
> >   of the R process).
> >   Hoping this can help,
>
> Yes, ESS uses something like that. But, Programming Perl, p.345, warns you
> not to use IPC::Open2 for this purpose, because of buffering, but Comm.pl
> (from CPAN) instead.  Emacs does use pseudo-ptys (on Unix, and tricks on
> Win32) to overcome the buffering problem. I am not sure that R always
> flushes its buffers, as although Rprintf seems to do so if it thinks it has
> an output file, R_WriteConsole does not.  (Crashes in batch jobs do
> sometimes seem to have incomplete output on the output file.)
>
> --
> Brian D. Ripley,                  ripley at 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-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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._



More information about the R-help mailing list