Variable scoping

There is an important aspect to understand with respect to variable scoping within loops. As an example, consider

  use strict;
  my $i = 12;
  print qq{Before the loop, \$i is $i\n};
  for (my $i=0; $i<3; $i++) {
    print qq{Within the loop, \$i is $i\n};
  }
  print qq{After the loop, \$i is $i\n};

This will print out

Before the loop, $i is 12
Within the loop, $i is 0
Within the loop, $i is 1
Within the loop, $i is 2
After the loop, $i is 12

What happens here is that the variable $i declared within the for loop has a lifetime only within that loop, and doesn't affect any possible variable $i declared outside of the loop. This also means that this program

  use strict;
  for (my $i=0; $i<3; $i++) {
    print qq{Within the loop, \$i is $i\n};
  }
  print qq{After the loop, \$i is $i\n};

will not run under use strict;, as the variable $i used in the print statement after the loop hasn't been declared within that scope.

While it might take some getting used to, variable scoping is a useful tool in debugging programs - although there are an effective infinite number of variable names one can use, we tend to use the same variable names for certian types of variables (for example, $i for an integer counting variable). Variable scoping can prevent a common error of using the same variable name in two places and assuming they're different. For example, consider

  $i = 1986;
  ..... # 1200 lines later
  for ($i=0; $i<4; $i++) {
    ......
  }
  ..... # 4300 lines later
  print qq{\$i is now $i};

In this example, $i is used in two places, presumably for a year (1986) and for a counter in the for loop. In the final print statement we might be expecting $i to be the year, but would be surprised to find it's value altered (through the for loop). Localizing the variable $i within the for loop would allow both variables $i to exist independently within their appropriate scopes.