different names for output files

rohr · 03-23-2005, 03:26 AM

Hi, i need help with the following:

i have two bash script, the first one produces as output ASCII data files and the second one takes these and produces ".eps" output graphs calling gnuplot to manage this.

Up to now, each time a call the first script and then the second i get, say, some graphical output files like fileA.eps, fileB.eps fileC.eps.

But if i make this procedure again, calling the two bash scripts over there, it overwrites fileA.eps, fileB.eps fileC.eps and i lost all ".eps's" i got the first time.

So i need some way such that if i call those bash scripts, say, four times, it gives me as output something like:

Code:

fileA1.eps fileB1.eps fileC1.eps
fileA2.eps fileB2.eps fileC2.eps
fileA3.eps fileB3.eps fileC3.eps
fileA4.eps fileB4.eps fileC4.eps

where each line correspond to a call for those two scripts.

Thank you in advance for any response.
Bye.

zeropash · 03-23-2005, 09:40 AM

Code:

#!/bin/bash

flist=`ls fileA*.eps 2> /dev/null`
fnum=`echo $flist | sed -e 's/.*\( \)*\(fileA\)\([0-9]\+\)\(\.eps\)/\3/g'`
#echo $fnum
if [ -z $fnum ]; then
  fnum=1
else
  fnum=`expr $fnum + 1`
fi
echo "creating fileA$fnum.eps"
touch fileA$fnum.eps
echo "creating fileB$fnum.eps"
touch fileB$fnum.eps
echo "creating fileC$fnum.eps"
touch fileC$fnum.eps

modify this as per your needs.

TheLinuxDuck · 03-23-2005, 09:44 AM

One way to handle this would be to initialize a counter variable at the beginning of the script. Then, run a while loop which tests for the existence of fileA$counter.eps, and increments the counter if it exists. Once you find an A file that doesn't exist, you can pretty safely assume that B & C don't exist either..

Code:

counter=1

while [ -e "./fileA${counter}.eps" ]; do
  counter=$(($counter+1))
done

# From now on, refer to files as fileA${counter}.eps, etc.
#

That should solve the duplicate issue.