[R] slow R start up

[Barry Rowlingson]

Some helpful person wrote:

> A possibility which might work, and if so would work as you intend,
> is to use a FIFO ("named pipe") or two to communicate with R.

>>>One possible solution may be to send the relevant commands to an
>>>already
>>>running copy of R but I've yet to figure out how to achieve this.

  You could try one of the R server-based solutions:

http://franklin.imgen.bcm.tmc.edu/R.web.servers/

  You could try Rserve, but you'll probably have to write some C++ 
wrapper so you can do a command from the shell, so you can do something 
like:

  echo "x=runif(1000);hist(x)" | RserveShell

from your shell script. Its probably only ten lines of C++.

http://stats.math.uni-augsburg.de/Rserve/doc.shtml

  If you rewrite your script in Python, I may have some Python-Rserve 
bindings working soon!

  Do you need to get any of R's computational results back into your shell?

  Whenever I write shell scripts, once it gets to about 20 lines I think 
to myself "Uh oh, I shoulda written this in perl". And when I've written 
200 lines of perl I find myself thinking "Uh oh, I shoulda written this 
in Python".

Baz