Hi, I have the following: File2: contain the following lines: a^ aaa^aa aa^~ b^ bbb^bb bb^~ c^ ccc^cc cc^~ d^ dddd^dd dd^~ File1: contain the following lines: b^ bbb^bb bb^~ c^ ccc^cc cc^~ I get File2 as input and I want to do as following: for each line in file2 { if (line exist in file1) print "!"line >> output else print "+"line >> output } I work on tcsh script: and this is what I wrote it: cat File2 | awk '{if (grep -c $0 file1 == 0) print "- "$0 >> output else print "! "$0 >> output Expected output: - a^ aaa^aa aa^~ ! b^ bbb^bb bb^~ ! c^ ccc^cc cc^~ - d^ dddd^dd dd^~ But I get: ! a^ aaa^aa aa^~ ! b^ bbb^bb bb^~ ! c^ ccc^cc cc^~ ! d^ dddd^dd dd^~ Do you have any idea what's wrong? Thanks, Roy. 4

You can run a shell command like grep from within awk, but you have to use the system function.  awk sees your grep command as several expressions: a non-existant variable named grep, etc.  So it is trying to compare a concatenated expression with zero, which in this case will always be false. I can show you the system call if you want, but a much better way to do this is to keep it all within awk: awk 'BEGIN { while ((getline < "file1") > 0)    file1[$0] = 1} {if ($0 in file1)     print "! " $0  else     print "- " $0 }' file2 > output