Isn't $i initialized with 0 inside the for loop?

it is... but if you test $i < @array and then access elements with an index $i*2, you're accessing elements outside of those you have defined...

Consider the different values that $i will assume as the loop runs. Potentially just print out $i like:

print "Slot[$i]: $array[$i*2], port: $array[$i*2 + 1]\n"
[download]

If you look at code written by Perl experts then you'll probably notice that they very rarely use the C-style for loop like you're using here. Most people find the shell-style foreach loop far easier to deal with.

In this case, I think you want:

foreach my $i (0 .. ($#array/2) - 1) {
  ...
}
[download]

But the set of people who avoid using C-style for loop are typically the ones that avoid $# as well.

I do use the C style for loop if I want the index, because, IMO, it's easier to avoid off-by-one errors than using foreach style. The latter means you either have to use '- 1', or remember to use '$#'.

for (my $i = 0; $i < @array; $i++) {...}
[download]

is standard idiom which translates easily between languages. Both

foreach my $i (0 .. @array - 1) {...}
[download]

and

foreach my $i (0 .. $#array) {...}
[download]

feel artificial, and Perl specific to me. And they both have their ugliness (either -1 or $#array).

foreach my $i (0 .. ($#array/2) - 1) { ... }

This produces an off-by-one error. Either
    ($#array/2)
or the uglier
    (@array/2) - 1
works as I understand the OPer requires.

>perl -wMstrict -le
"my @ra = qw(a A b B c C d D e E);
 for my $i (0 .. ($#ra/2) - 1) {
   print qq{$ra[ $i * 2 ] : $ra[ $i * 2 + 1 ]};
   }
"
a : A
b : B
c : C
d : D

>perl -wMstrict -le
"my @ra = qw(a A b B c C d D e E);
 for my $i (0 .. ($#ra/2)) {
   print qq{$ra[ $i * 2 ] : $ra[ $i * 2 + 1 ]};
   }
"
a : A
b : B
c : C
d : D
e : E
[download]