hello phizymonk, welcome to the monastery and to the wonderful world of Perl

following the path GrandFather drawn for you (think about the program before coding) and as you are learning Perl i propose you some pseudo-code insetead of a a full featured script that satisfy your task. I add some sparse suggestion that you can try to expand to code. GrandFather insisted on specifications: how the user will interact with the program? it is important! Generally speaking if you are able to write in plain language the solution, then you'll be able to code it or at least (if your skill are not enought) to spot what part you are not able to code: it is a great advantage.

As foreword suggestions, try to code in a plain and well documented way, idiomatically: as already said by GrandFather use strict and warnings, use open with 3 arguments (idiomatic = more reliable and less error prone). Try to isolate behaviours in subroutines: it will help a lot when the code become more complex. And, very important, choose meaningfull names for your variables.

I think you simply need an array instead of an hash: you definetively read an array of lines from file. Work with this array; it is indexed, it is easy to use and faster and without side effects (duplicate can exists in a file and in array not in an hash)

# pseudocode
# PART 0 - load useful stuffs and define some useful vars
use strict;
use warnings;
# maybe use Term::ReadLine; # a useful tool to grab user input


my $filename = $ARGV[0] || 'defaultfile.txt';
my $separator = ' - '; # in your OP you split using '-' and this leave
+s a blank space after the item an another before the price!
my $was_modified = 0;  # count total modifications

make a copy of the original file with a date in the name and possibly 
+another exstension, just in case

open the file for reading
my @alllines = <$file_handle>; # in list context <> reads all the line
+s (good for not huge files)
                               # ATTENTION: you still have all newline
+s at end of each record!!
close file_handle or die



# PART 1 - present the file to the user: show also the index of the ar
+ray that we will use to modify the entry, later
# better to define a sub to show the file content:

sub showfile {
 foreach my $index( 0 .. $#alllines  ) {  # $#alllines is the last ind
+ex (the number of) of the @alllines array

  chomp $alllines[$index];  # remove newlines
  
  print "[$index] $alllines[$index]\n"

 }
 modify_from_input; # call teh sub defined below: grab user input and 
+come back to this sub again
}

showfile; # call the sub



# PART 2 - edit prices
# grab the user input and modify the array entry. a sub is better

sub modify_from_input {
 print "Enter the index you want to modify: (0 - $#alllines)\n";
 my $input = ... # grab the input
 chomp $input;   # remember newlines!
 
 if input is invalid warn something and call the same sub again
 else
  my ($name, $price) split $separator, $alllines[$index];
  print "Enter new price for [$name] (was [$price]) or 'write' to writ
+e the file:\n"
  my $new_price = <STDIN>;
  chomp $new_price;
  # validate $new_price ?
  
  if new_price is empty -> remove the item? -> splice the array -> sho
+wfile
  return if $new_price eq 'write'; # break the loop of the 2 subs and 
+go to the end of program
  # modification!
  $was_modified++; 
  $alllines[$index] = $name.$separator.$new_price;

  showfile; # we go back in show mode
}




# PART 3 - modify the file: we arrive here only if user enter 'write' 
+while in the modify_from_input sub 


open my $write, '>', filename or die  # here we are rewriting the orig
+inal file
print $write "$_\n" for @alllines;
if $was_modified print "$was_modified modifications were applied\n"
else print "no modifications\n"
close $write or die;
print "$filename succesfully written\n";
exit 0;
[download]

have fun filling the blanks!

L*