From: Josh Cheek <josh.cheek@...>
Date: 2009-10-21T05:49:48+09:00
Subject: Re: Randomly generated words from a set of letters

--000e0cd2be98b12142047664002e
Content-Type: text/plain; charset=ISO-8859-1

On Tue, Oct 20, 2009 at 10:26 AM, Ilan Berci <ilan.berci@gmail.com> wrote:

> unknown wrote:
>
> >
> > 1) Manually add the pre-defined word set to a DB table and randomly pull
> > a word out as needed.
> >
> > 2) Each time I need a word, build a random word from the set of letters
> > and compare to a dictionary somewhere to make sure I have a real English
> > word.
> >
> > 3) Build a complete random word set one time (as in 1 but with all
> > possible word combinations) and store in the DB to pull from randomly.
> >
> >
> > So...what would you do?  1 is probably easiest....but 2 is more
> > interesting for me from a problem solving standpoint.
> >
> > Just looking for some thoughts.
> >
> > Thanks,
> > Matt
>
> Not sure what the difference is between 1 and 3 but option number 2 can
> be grossly inefficient for little or no benefit.
>
> Option 3 is O(1) complexity and therefore I vote for it..
>
> hth
> ilan
>
> --
> Posted via http://www.ruby-forum.com/.
>
>
What if you created some sort of a tree data structure, where each node
represented a letter. (each node has at most 26 children) That seems like it
should be more efficient, because for something like toon , tool , took ,
tooth , they all utilize the same series of paths to go through the too, and
only diverge after that. So instead of 17 characters required for these 4
words, you only need 8.

--000e0cd2be98b12142047664002e--