RE: Automation and Preprocessors

Hi Ali,

For repetitive tests where you modify the numerical value of some
parameter (e.g. some component in the geometry), bash (or equivalent) is
a good option (while excel probably isn't).

For example, if you want to send simulations with different diameters in
the body basII of an input file called 'start.inp', you could use something like:

#!/bin/bash
i=5
while [ $i -le 9 ]
do
cat start.inp |
sed "s/RCC basII 15.24 -48.26 -9000.0 0.0 0.0 205.74 5.0/RCC basII 15.24 -48.26 -9000.0 0.0 0.0 205.74 ${i}.0/" > file$i.inp
i=`expr $i + 1`
$FLUPRO/flutil/rfluka -e ./flukahp -N 0 -M 1 file$i
done
exit 0

Revise basic documentation of bash commands. With relatively easy
commands you could an do more complicated schemes where you would parse
results from FLUKA and based on those modify your input in one way or
another in the subsequent runs.

Mario

-----Original Message-----
From: owner-fluka-discuss_at_mi.infn.it on behalf of Ali Koosha
Sent: Sun 4/18/2010 8:11 AM
To: fluka-discuss_at_fluka.org
Subject: Automation and Preprocessors

Hi all,

I'm going to do a large amount of progressive simulations were some input
parameters
(e.g. geometry, energy of the source particles, etc) could be changed
according to a predefined scheme.

Could anyone help me to incorporate some sort of automation into my project,
to minimize the
required distinct stages for input making, running and, output readings? I'm
not familiar with the Fluka pre-processors, but it seems that the literature
is free of valuable docs.

Cheers
---------
A. Koosha
Received on Tue Apr 20 2010 - 00:13:41 CEST