I'm taking a guess here since you don't give us an example of what output.txt should look like, but I am guessing it is in this part:

while (<FILE>){
  /^(.*?): \s*(.*)$/;    # Save text before colon in $1, and after 
                         # colon in $2 - Taken from the Camel Chp. 5.9
  $fields{$1} = $2;      # Creat %fields hash

}
[download]

You are not checking to see if the RE matches before trying to use $1 and $2. You should probably have something like:

while (<FILE>){
  if (/^(.*?): \s*(.*)$/) {    # Save text before colon in $1, and aft
+er 
                               # colon in $2 - Taken from the Camel Ch
+p. 5.9
    $fields{$1} = $2;          # Creat %fields hash
  }
}
[download]

Update: As AM pointed out below, you may even need more error checking depending on what your output.txt file looks like. Another Update: If I make the change I pointed out above to the code, and run it against the output.txt you posted below, I do not get an error message.