PWC158 - First Series Cuban Primes


On with TASK #2 from The Weekly Challenge #158. Enjoy!

The challenge

Write a script to compute first series Cuban Primes <= 1000. Please refer wikipedia page for more informations.


7, 19, 37, 61, 127, 271, 331, 397, 547, 631, 919.

The questions

Where’s the trick? I feel I’m missing something important here.

The solution

I could think of two ways to address this challenge:

  • go through prime numbers up to 1000 and find out those that are also of the form $3y^2 + 3y + 1$, OR
  • generate all numbers of the form $3y^2 + 3y + 1$ up to 1000, and find out those that are also primes.

Eventually I settled for the latter, because generating the numbers is easier than finding the integer solutions to the quadratic eqution, and the test for primality is readily available in ntheory.

So let’s start with Perl:

#!/usr/bin/env perl
use v5.24;
use FindBin '$Bin';
use lib "$Bin/local/lib/perl5";
use ntheory 'is_prime';

my $M = shift // 1000;
my @cubans;
my $y = 1;
while ((my $p = 3 * $y * ($y + 1) + 1) <= $M) {
   push @cubans, $p if is_prime($p);
say join(', ', @cubans), '.';

I’m not entirely happy with this solution, it feels… clunky. Like collecting the stuff in array @cubans, I don’t know.

The Raku counterpart allows me to use one of my favorite constructs, i.e. gather/take:

#!/usr/bin/env raku
use v6;
sub MAIN (Int:D $M = 1000) {
   put gather {
      my $y = 1;
      while (my $p = 3 * $y * ($y + 1) + 1) <= $M {
         take $p if $;
   }.join(', '), '.';

I don’t know why I’m so fond of gather/take, it just feels so natural. Before you get spoiled too, though, keep in mind that I’be been advised against it because of performance issues. This is one reson why I like these challenges, I can play with the language without being hit by it! 😅

Stay safe!

Comments? Octodon, Twitter, GitHub, Reddit, or drop me a line!