Shell Programming and Scripting

Hello,

Im having some difficulty writing a krn shell script

The script must accept a text file with 3 columns
Take each column to be a different variable
then, take each line in the first column as a string, and replace a similar string in a filename in a different directory with the content of the second column.

The script must then append the third column onto the front of the filename changed (and put a period in between the filename, and the column data).

I wrote this trying to use several awk's for accepting columns, but was unsuccessful.

Any help is dually appreciated.

Are the files text and is any match OK anywhere on any line any number of times? No '/' in the data?

Code:

while read a b c
do
 sed 's/'"$a"'/'"$b"'/g' file2 >$c.file2
done <file

I tried to work that in, but I suppose my issue was a little different than I explained it, or can be solved more logically than I tried to solve it.

I have a file with 3 columns like so....

xxxxxx xxxxx xxxxxxx
xxxxxx xxxxx xxxxxxx

spacing in between the columns is indeterminate. I need to run through each line of the file, which I can do using a while loop. However, after that, I need to be able to use the first second and third strings in the file as variables.

I then need to grep a directory for the first string, place the second string onto the front end of each filename the first string is found in, and replace the first string with the third string in the original file. I tried to do this using an awk, but was unsuccessful, although I would like to use awk rather than sed for speed.

---------- Post updated at 11:33 AM ---------- Previous update was at 09:43 AM ----------

Maybe I should start smaller,

I have an awk command to output columns as $1 $2 $3
is there any way to take the output $1 and have a grep command to search for a match of any string which matches a string in $1 in a directory

so... $1 is a column

xxxxxx
xxxxxx
xxxxxx
xxxxxx


and i want to search for each individual string

xxxxxx

in a directory, and then replace each matching string, with a string from a different column, on the same row.