Re^3: Refactoring just to refactor?

[stevieb]

A subroutine will return to its caller the result of the last evaluated expression. I would recommend to use return() in all cases within all branches, unless you know why/when not to.

Here's an example of output of values. Note that say() is just a glorified print(), but it adds a newline at the end by default. Nothing special here...

use warnings;
use strict;
use feature 'say';

sub return_print {
    my $x = 254;
    print "$x\n";
}
sub return_return {
    my $x = 255;
    return $x;
}

say "print:  " . return_print();
say "return: " . return_return();

Output:

254
print:  1
return: 255

So, the first output (254) is the output from the print() call. The second output (print: 1) is the output of a successful call to print(), which is 1. The third value is the return value explicitly sent via return().

I would highly recommend you play around and get a grasp on what return values you're getting from and/or expecting from your function/method calls.

All of your tests should be checking against all possible and potential return values so there aren't any unexpected results. If there are any odd things found, add more tests, or fix your subroutines to ensure they only return very specific value(s).