Question

There is an unload.dat file with the list of .ksh file names. The task is to extract a few parameter details from each of the files that are present in the unload.dat file. Every file contains the below information.

Let us consider for first file name (file_name1.ksh) in the unload.dat:

graph_name = abcd.ksh

export name_a=value_a
export "name_b=value_b"
export name_c=value_c
export name_d=value_d
export name_e="value_e"
export name_f="value_f"

The next file’s fetched data should be appended to the previous file’s output in the same output file.

The output should look like the below:

abcd.ksh    name_a  value_a 1   filename_1.ksh
abcd.ksh    name_b  value_b 1   filename_1.ksh
abcd.ksh    name_c  value_c 1   filename_1.ksh
abcd.ksh    name_d  value_d 1   filename_1.ksh
abcd.ksh    name_e  value_e 1   filename_1.ksh


abcd.ksh    name_a  value_a 2   filename_2.ksh
abcd.ksh    name_b  value_b 2   filename_2.ksh
abcd.ksh    name_c  value_c 2   filename_2.ksh
abcd.ksh    name_d  value_d 2   filename_2.ksh
abcd.ksh    name_e  value_e 2   filename_2.ksh

Answer 1

#!/bin/sh

counter=1
while read -r filename; do
  if [ "$counter" -ne 1 ]; then
    echo
    echo
  fi

  IFS== read _ graph < "$filename"
  graph=${graph# }

  sed -n '/^export/{s/^[^ ]* //;s/"//g;p}' < "$filename" \
  | while IFS== read -r name value; do
    echo "$graph    $name  $value $counter   $filename"
  done

  counter=$((counter+1))
done < unload.dat