Perl subroutine sort my_compare (@d);

Question

I am taking an online perl class and we started subroutines. The professor used this example:

@d = (1, 5, 2, 9, 8, 7, 10, 11);

@c = sort my_compare (@d);

# Print the result
print "Before sort:\t".join(", ", @d) . "\n";
print "After sort:\t".join(", ", @c) . "\n";    

sub my_compare
{
    if ($a < $b) { return -1; }
    elsif ($a > $b) { return 1; }
    else  { return 0; }
}

I dont understand the @c = sort my_compare (@d); statement and why I cannot rename the variables $a and $b . This is completely different than what I am used to doing for soubroutines in languages like C and C++.

Could somebody explain this to me?

Thank you for your help! Zahra

Answer 1

Unlike many other languages, Perl is like slang. Less consistent, but full of very useful shortcuts. As Larry Wall stated:

In general, they (Perl operators) do what you want, unless you want consistency.

For the C library qsort function, you always need to provide a function pointer called compar , that will do the actual comparison of value pairs.

void qsort(void *base, size_t nmemb, size_t size,
           int (*compar)(const void *, const void *));

This is very similar in Perl, but providing comparison code is optional. This uses the default built-in alphabetic comparison.

sort @d

The whole reason of the above exercise is to replace the alphabetic comparison with a numeric one. You can provide custom comparison code in the form of a subroutine name or a block.

sort subname @d;
sort { $a <=> $b } @d;

The above is the most simple solution to your problem, using the numeric comparison operator <=> . It does the same as sub my_compare .

And well, $a and $b are just built in. Perl has no function prototypes with parameter names, so they had to name it somehow.

See http://perldoc.perl.org/functions/sort.html for further details.