Read and Replace

Read and Replace

--------------------------------------------------------------------------------

Hi!
I think my last post was a little confusing, so I will try an post a sample this time.
I have a script which is supposed to output two columns of a file to a different file, then take the first column, search a directory for it, and replace anything it finds with the second column. This is what I have:

for file in /XXXX/XXXX/XXXX/I
do
echo $VAR|awk '{printf "%d\n",$1}'|read NEWVAR
done
awk < XXXX_table.txt '{print $1 $3}' >> var1
for line in /XXXX/XXXX/XXXX/var1
do
x=$(awk < var1 '{print $1}')
y=$(awk < var1 '{print $2}')
cd /XXXX/XXXX/XXXX/I
sed "s/$x/$y/g"
echo $x
echo $y
done

The first loop is supposed to remove all of the leading zeroes in all the files in that directory as well, but I guess that is wrong too.

Having lots of difficulty on this one, but I feel as though the answer is very simple

First problem - where does $VAR come from?
Why are you doing that first loop which will only (once it runs correctly) ever produce a result that you can use from the last iteration.

Let's go back to basics. You have a directory, containing files whose names follow a particular pattern. Show a couple of examples of the names. You mention the files have two columns which you wish to do some operations upon. Show a couple of examples of the content. Also, do the files contain just one line or many? If many do you want the operation to be performed on all or just some of those lines?

Don't use awk < a_file_name '{print $1}' - use: awk '{print $1}' a_file_name

You refer to file names ($VAR and XXXX_table.txt) without a path, then do a change directory (in a loop, which is a redundant thing to do!).