in reply to What about Easter?

You'll love php for this: http://www.php.net/manual/en/function.easter-date.php

Replies are listed 'Best First'.
Re^2: What about Easter?
by pg (Canon) on Jul 24, 2005 at 17:24 UTC

    Okay I ported whatever is in the last post of that thread to Perl: (only tested with this year though)

    print easter_date(2005); sub easter_date { my $Year = shift; $G = $Year % 19; $C = int($Year / 100); $H = int($C - ($C / 4) - ((8*$C+13) / 25) + 19*$G + 15) % 30; + $I = $H - int($H / 28)*(1 - int($H / 28)*int(29 / ($H + 1))*( +int(21 - $G) / 11)); $J = ($Year + int($Year/4) + $I + 2 - $C + int($C/4)) % 7; $L = $I - $J; $m = 3 + int(($L + 40) / 44); $d = $L + 28 - 31 * (int($m / 4)); $y = $Year; return sprintf("%04d-%02d-%02d", $y, $m, $d); }

      DON'T USE THIS CODE!

      I wrote a quick and dirty compare of the date calculated by this sub against the list in the zip file at Easter Sunday Dates and found it to be 22% inaccurate. Good enough for PHP perhaps? *grin*

      Yes I get that the thread was a joke :)

      Update: see Re^3: What about Easter? by themage for an updated subroutine which does not produce any erros when compared to the list.

      --
      Murray Barton
      Do not seek to follow in the footsteps of the wise. Seek what they sought. -Basho

      There is a small error in the PHP version, that pg ported here: <updated, wrong name>.

      The line
      $H = int($C - ($C / 4) - ((8*$C+13) / 25) + 19*$G + 15) % 30;

      Should be:
      $H = ($C - int($C / 4) - int((8*$C+13) / 25) + 19*$G + 15) % 30;

      The final sub sould look like:

      sub easter_date { my $Year = shift; my ($G,$C,$H,$I,$J,$L)="0"x6; $G = $Year % 19; $C = int($Year / 100); $H = ($C - int($C / 4) - int((8*$C+13) / 25) + 19*$G + 15) % 30; $I = $H - int($H / 28)*(1 - int($H / 28)*int(29 / ($H + 1))*(int(21 +- $G) / 11)); $J = ($Year + int($Year/4) + $I + 2 - $C + int($C/4)) % 7; $L = $I - $J; my $m = 3 + int(($L + 40) / 44); my $d = $L + 28 - 31 * (int($m / 4)); my $y = $Year; return sprintf("%04d-%02d-%02d", $y, $m, $d); }
      This way, I think that it should be ok. What you think greenFox?

        That fixed it. With your updated sub I got 0 errors for 2517 dates. Well spotted.

        --
        Murray Barton
        Do not seek to follow in the footsteps of the wise. Seek what they sought. -Basho

      Too SLOW! Must be in CORE!
Re^2: What about Easter?
by biosysadmin (Deacon) on Jul 24, 2005 at 21:48 UTC
    The slight chuckle that I had after reading the original post was swept away by the shocking realization that someone, somewhere considered this necessary enough to be in the core language. Wow.