Hi, I want to excute a shell script by passing a CSV file as argument, where the file name will be based on the file that is been placed in a particular directory. Ex: cleaner.sh GSM20050609.csv the argument file will be the one in the directory. How to read the file name and pass it. Thanks Siva Hi, I'm trying to replace a column in a comma seperated file with a column in another file using sed/awk. Can anyone help me. Thanks Siva

if [ $# -ne 1 ] ; then    echo '\nUsage: cleaner.sh <filename>\n'    exit 1 fi dir=/home/particular fname=$1 if [ ! -r $dir/$fname ] ; then    echo "$fname does not exist in $dir"    echo "   or I have no read access"    exit 1 fi echo "Now processing $dir/$1 ..."

Jim, Think my question is not clear, let me put it in other way.. In /home/files I have 2 files GMS.txt and AMS.txt I've my script /home/scripts/cleaner.sh Now I want to pass all the files in /home/files to cleaner.sh as argument one by one. like cleaner.sh GMS.txt cleaner.sh AMS.txt I want a script which will get files from /home/files directory and pass it to cleaner.sh as argument. Can you help me. Thanks Siva Hi, I'm trying to replace a column in a comma seperated file with a column in another file using sed/awk. Can anyone help me. Thanks Siva

Sure, here's a couple of different ways: find /home/files -exec c /home/scripts/cleaner.sh {} \; or for fname in $(ls /home/files) do /home/scripts/cleaner.sh $fname done On the find command approach, you can add additional filters to the find command, such as: regular files only (ignore directories etc) modify date filter file size filter file ownership etc

Jim, Yeah....it works great. Thanks Siva Hi, I'm trying to replace a column in a comma seperated file with a column in another file using sed/awk. Can anyone help me. Thanks Siva