#!/usr/bin/env bash

outfile="$1.filtered"

# Only print the lines we want
filter_dat() {
  awk 'BEGIN { RS = "" ; FS = "\n" }
    {
      if ( $2 !~ /(text|error|note|title)/)
      { 
        print $1
        print $2
        print $3
        if ($4 !~ /UTF-8/) print $4
        if ($5 !~ /UTF-8/) print $5
        if ($6 !~ /UTF-8/) print $6
        if ($7 !~ /UTF-8/) print $7
      }
    }' $1
}

# Run filter_dat then pipe output to sed to strip all empty
# lines, then add one newline above Name and save it to a file
filter_dat $1 | sed '/^$/d; s/Name/\n&/g' > "$outfile"

printf '%s%s\n' 'Processing complete! The filtered file is: ' "$outfile"