Loops with Self-Modifying Code

Today, we're exploring an unconventional way to create a loop in Perl using more self-modifying code. This technique, while not practical for everyday use, showcases Perl's flexibility and power. Let's dive into this intriguing code snippet:

​x
#!/usr/bin/env perl

use feature qw|say|;

$_ = <<'EOF';
10; 
s~(\d+)(?{ say qq($1) })~$1-1~e;
sleep 1;
$1 ? eval : say q(Countdown complete!);
EOF

eval;

This code creates a countdown from 10 to 1 without using any traditional loop constructs. Let's break it down:

We start by assigning a heredoc to the $_ variable. This heredoc contains the code that will be repeatedly evaluated.

The first line in the heredoc sets our starting number: 10.

The substitution operator s~~~ is the heart of our loop simulation:

xxxxxxxxxx
s~(\d+)(?{ say qq($1) })~$1-1~e;

This line does several things:

• It matches a number (\d+)

• It prints the matched number using a code block (?{ say qq($1) })

• It replaces the number with itself minus one ($1-1)

• The 'e' flag at the end tells Perl to evaluate the replacement as code

We add a sleep 1; to slow down the countdown for visual effect.

The final line in the heredoc is our exit condition:

xxxxxxxxxx
$1 ? eval : say q(Countdown complete!);

If $1 (our number) is truthy (non-zero), it evaluates the code again. Otherwise, it prints "Countdown complete!"

Finally, outside the heredoc, we have a single eval; statement. This kicks off the whole process by evaluating the code in $_.

When run, this script will count down from 10 to 1, pausing for a second between each number, and then print "Countdown complete!"

While this example is more of a curiosity than a practical coding technique, it illustrates Perl's flexibility and the creative ways we can bend the language to our will. It's a reminder of why Perl is often described as a "Swiss Army chainsaw" of programming languages!