[FIXED] awk script inside jenkins pipeline

Issue

I am trying to execute below script inside jenkinsfile

awk 'NR==FNR{new=new $0 ORS; next} /^}]$/{printf "}],\n%s", new} 1' output2 ${WORKSPACE}/${REPO}/file1 >> output2

This script works fine in linux command promt but it gives error when tried to run inside jenkins file.I tried below

sh """
   awk 'NR==FNR{new=new $0 ORS; next} /^}]$/{printf "}],\n%s", new} 1' output2 ${WORKSPACE}/${REPO}/file1 >> output2
"""

But it gives syntax error

Solution

You need to watch out for escaping single or double quotes, backslashes and dollar signs. Here I've escaped everything:

sh """
   awk \'NR==FNR{new=new \$0 ORS; next} /^}]\$/{printf \"}],\\n%s\", new} 1\' output2 \${WORKSPACE}/\${REPO}/file1 >> output2
"""

Maybe you actually don't want to escape ${WORKSPACE} or ${REPO} if those are pipeline environment variables that you want substituted by groovy as opposed to by sh. And maybe you don't need to escape the single quotes inside a triple """ multistring. But you definitely need to escape the double quotes and the backslash in \\n.

Answered By - david_beauchamp
Answer Checked By - Mildred Charles (JavaFixing Admin)