-- Randal L. Schwartz, Perl hacker
Be sure to read my standard disclaimer if this is a reply.

Are you doing the sort on numeric data or strings? If so, you need to use '==' instead of 'eq', '!=' instead of 'ne', and '<' insetad of 'lt'. Or was that just part of the psedocode style?

Also, your comments seem to indicate you're showing how to do a sort in Perl for academic reasons. If not, you should just use sort instead.

----
Invent a rounder wheel.

($array [$element_one], $array [$element_two]) = 
    ($array [$element_two], $array [$element_one])
[download]

update

It is certainly not a "classic" bubble-sort but a curious mix of elements of an exchange sort (namely the swapping of the current record with the smallest remaining record) and an inverted type of insertion sort (namely finding the next smallest record and putting it directly in its place)

None of these sorts are known to be fast or efficient and by combining them you probably get the worst elements of both.

If this is homework, it is bad homework. I would never suggest a student to implement such a bad algorithm.

To know more about sorting in general, one could do worse than read Don Knuth's "The Art of Computer Programming - Vol. 3 - Sorting and Searching".

CountZero

"If you have four groups working on a compiler, you'll get a 4-pass compiler." - Conway's Law

You say you could do this in Java. I suggest you write it in Java. When it's working, translate the java syntax into perl as far as you can. When you get stuck, post the Java code and your attempted translation hilighting the lines your having trouble with. You might then get assistance in correcting those lines.

It's good to understand how things work, and sorts are often used for this purpose when learning.

It seems to me that you want to do this in a perlish way, and that is why you are trying to do it with a foreach loop. However, foreach (@entries) is the wrong type of loop to use for sorting @entries. It can be done, but you're trying to pound a square peg into a round hole. Why? foreach goes through each element of an array, but it doesn't tell you where in the array the elements are located. A sorting routine, by its nature, needs to have at least some information about the location in the array of the elements it's comparing.

It's a good idea to use correct names for your variables, lest you confuse yourself. The loop variable in a foreach loop is not a counter, unless the array you're iterating over happens to be a counting sequence.

Thank you for your help. I WAS trying to find a way to use the power of the foreach structure, but you showed me the error of my ways. Would it be better to say something like this:

for (i = 0, i =< @#entries, i++)
     {
     $indexofleast = i;
     for (j = 1, j =< @#entries, j++)
          {
          if (@entries[j] lt @entries[i])
               {
               $indexofleast = j;
               }
          if ($indexofleast != i)
               {
               $temp = @array[j];
               $@array[j] = $@array[i];
               $@array[i] = $temp;
                }
          }
     }
[download]

Thanks for anymore help I can get. I stress that I am brand new to perl (so I'm really not sure if I'm using the @# operator correctly). This is supposed to work on single character strings and sort them.

Close. I think you want $#entries, $array[i] and $array[j].

sub mysort {
    chomp(my @entries = <STDIN>);
    return sort @entries;
}
[download]

I know that Perl is supposed to make the routine easy, but I'm specifically supposed to implement a sort subroutine without using any built in perl functions.

Get thee to your nearest bookstore and get a copy of the wolf book (ISBN ISBN 1565923987), which has an entire chapter on sorting in perl without resorting to sort().

If you have to re-invent the wheel, then please do yourself a favour and use some more efficient algorithm, e.g. a Quicksort or a Shell-sort.

CountZero

"If you have four groups working on a compiler, you'll get a 4-pass compiler." - Conway's Law