Hmmm... let's see if I can explain this clearly. Basically, the code you originally posted spit out data as soon as it encountered a unique line. That is, it kept track of the previous line and as soon as the current line was different from the previous one (unless($line eq $oldline)), it would print it. The problem with printing the new unique line immediately is that you cannot count how often it appears in sequence, because you don't know yet. You've only found the first occurance of it at that point.

The solution then is to print each unique line after its last appearance in a sequence, not the first. To answer your question, that's what the "else" part within the loop does. It does not stop the loop (which just iterates over all of your input lines, after all).

The else executes if the current line is different from the one that's being kept track of (the $oldline). In other words, it executes as soon as you encounter a new unique line. The code prints the line along with the count, and then resets the counter (the $n = 1; line).

That last if statement will actually always evaluate to true. It says: "If there is any text in $oldline, then print it along with its count. If we didn't do this, the very last unique line would never get printed.

To pseudo-code it:

Read in my first input line and store it in $oldline
Now loop over the rest of the input lines
   if the current line is the same as $oldline, then
      increment our count
   otherwise (that is, the current line is different)
      print $oldline and the counter
      reset the counter
      set the current line as the next line to keep track of by storin
+g it in $oldline
Loop
if there's something in $oldline (which will always be the case), prin
+t it
[download]

It just struck me... I hope I haven't just done your homework assignment or something.

Update: I see it was homework, but in light of imhotep's explanation and the disclaimers on previous nodes, I don't mind having done it. I miss college anyway.