OpenEuphoria

Re: contest #2

• Posted by Derek Parnell <ddparnell at bigpond.com> Mar 17, 2002
• 339 views

Didn't Robert say something about posting actual code is not allowed.

I don't hash the dictionary words, just their patterns. Also, I don't use an
algorithm that looks
anything like Euman's. The "alphabet" that I work with is thus in the range 1 -
32. My approach is
to pre-calculate as much as possible at start up, so that inside the inner
loops, I don't do any
unnecessary calcs. Also, I try to avoid 'expensive' compuations like multiply
and divide and nested
sequence references. And append() is almost totally avoided, but when I do have
to, I append more
than I need so that in the next few iterations I can avoid doing another append.

I organise the dicionary by pattern. Note that over 95% of the patterns in the
dictionary have less
than 20 words in them, and only a few have more than 2000 words.

----------
Derek.


18/03/2002 10:57:57 AM, "C. K. Lester" <cklester at yahoo.com> wrote:

>
>
>euman at bellsouth.net wrote:
>> 
>> I'd like to see Derek & Eumans Hash Function take
>> the winners circle..
>
>Okay, Derek or Euman, how about posting your latest hash algorithm?
>
>Thanks!
>ck
>
>
>
>
---------
Cheers,
Derek Parnell 
ICQ# 7647806