Shell Programming and Scripting

Hi everyone ! I have a script to catch the same content of 2 files like below:

Code:

#!/bin/bash
COUNT=`wc -l file1`
LINE=(${COUNT[@]})
for i in `eval echo {1..$LINE}`
do
    y=`sed -n "$[i]p" file1`
    grep -e ^"$y"$ file2 >> file3
done
exit

Problems is in file1 and file2 there're line with [ ] characters and grep interpret them to regex. So they not appear in file3. Can anyone help me solve this ?

Try this:

Code:

while read line
do
if [ `grep -c "$line" file1` -gt 0 ]; then
   echo "$line" >> file3
fi
done < file2

here file3 should have the contents common in both file1 and file2.

cheers,
Devaraj Takhellambam

It works !! Thank devtakh !!

IMO this is not enough. To obtain a strict line match one needs to use fgrep or some special characters will still be interpreted. Also setting IFS to "" and using read -r ensure that the line that is read remains intact. Furthermore the -x option specifies a full line match. Also the use of printf instead of echo makes sure that \ in the character string does not get interpreted as an escape, which is the default on some systems. Lastly you can do a redirect from the loop body so you do not need to use append and remove the file beforehand. Finally, you can use grep's return code instead of its count so you do not need a test nor a sub process:

Code:

while IFS= read -r line
do
  if fgrep -xq "$line" file1 
  then
    printf "$line\n"
  fi
done < file2 > file3

But this means you can also do this:

Code:

fgrep -xf file2 file1 > file3

-or-

Code:

fgrep -xf file1 file2 > file3

Thank you very much Scrutinizer !! I get it clearly