in reply to Re^2: Felling a tree JAPH
in thread Felling a tree JAPH

For a given version, I get a consistent result. I reckon the discrepancies between different version are due to internal changes to the GC.

Relying on DESTROY is problematic, because the order in which thingies are GC’d is unspecified.

Replies are listed 'Best First'.
Re^4: Felling a tree JAPH
by MidLifeXis (Monsignor) on Sep 04, 2013 at 11:53 UTC

    That is true. Since they are consistent per version (I wonder if per platform makes any difference), I should see if I can reorder the data to make it work out on each :-)

    --MidLifeXis

[reply]
      What about making the letters depend on the following one? 
      sub DESTROY { print $_[0][0] }
$p = bless [$_, $p] for reverse split //, 'Just another Perl hacker,';

      [download]
       Update: Does not work in a different version. Ignore.
      لսႽ† ᥲᥒ⚪⟊Ⴙᘓᖇ Ꮅᘓᖇ⎱ Ⴙᥲ𝇋ƙᘓᖇ
[reply]
[d/l]

        Sounds very similar to the concept behind Destructive JAPH. I wonder if it would help solve the ordering issue on the destroy.

        These aren't really obfuscated, as much as exploring some dark (and some not so dark) corners of the language. I wonder if there is a more appropriate place for these, just so I don't get drawn into obfuscating them and masking the real thing I am playing with.

        --MidLifeXis

[reply]

In Section Obfuscated Code