OpenEuphoria

Re: Spell checker optimization

• Posted by Joe Phillips <bubba at TXWES.EDU> Mar 25, 1998
• 825 views

At 11:00 AM 3/25/98 -0600, you wrote:
>Well, I'm not sure I agree with Mr. Nieuwenhuijsen's idea of the
>binary tree implementation, but I'll give him a chance to explain
>it more fully before I say it's not a better way of doing it.

Thanks for the detailed explanations. Although I knew the concepts, the
presentation was succinct and thorough.

In fact, it was those explanations that leads me to think a Binary Tree
might be the best for this implementation.  Besides, what good is a
dictionary to which you cannot add words?  I think the b2-tree (balanced
binary tree) concept works beautifully.  In fact, the file can be stored as
a b2-tree (with pointers) so as to not have to rebuild the tree at each load.

Although, I often use binary searches for the very reasons mentioned.
(EASY) In fact, many times I will even use binary searches in un-sorted,
linked lists or files.  Although I have to parse through pointers, key
comparisons are kept at a minimum.



--[ Joe Phillips, Sytems Analyst
--[ Texas Wesleyan University     817-531-4444
--[
--[  "The tongue of the just is as choice silver:
--[   the heart of the wicked is little worth."  Proverbs 10:20