UNIX for Dummies Questions & Answers

Hi,
I am trying to write a shell script designed to take input line by line by line from a file with a word on each line for editing with sed. Example file:

1.ejverything
2.bllown
3.maikling
4.manegement
5.existjing
6.systems

My design currently takes input from the user, and stores it in a variable (linenum) which is then used in sed as the line number. I have,


echo "Which line number"
read linenum

And then use $linenum in sed as the line number in the file, ex

sed '"$linenum"i\
> FINDSTRING\
> REPLACESTRING
> ' foo > foo.old

I haven't got the above to work so far. I guess my question is,
can you use variables in a sed command line? If so how?

I also saw the option of using a loop to accomplish the job somewhere
as shown below...

#!/bin/bash
for in $fl; do
mv $fl $fl.old
sed 's/FINDSTRING/REPLACESTRING/g' $fl.old > $fl
rm -f $fl.old
done

Please help!

My apologies, it was not my intention to duplicate my thread, I couldn't find a way to move my thread to another sub-forum. Any suggestions?

if you wanna hardcore the replace string you can use this awk one liner

Code:

awk -v var=$linenum /$linenum'\./{gsub($0,var".REPLACESTRING")};{print}' input_file

otherwise take input from user for replace string and use below command

Code:

awk -v var=$linenum"."$replacestring /$linenum'\./{gsub($0,var)};{print}' input_file