From: Chris Hulan <chris.hulan@...>
Date: 2010-09-29T03:10:12+09:00
Subject: Re: levenshtein function issue

On Sep 27, 3:37 pm, Joe Smith <macsare....@gmail.com> wrote:
> Chris Hulan wrote:
> > On Sep 26, 12:43 am, Joe Smith <macsare....@gmail.com> wrote:
> >> error... (that's why I had originally changed it)
> >> --
> >> Posted viahttp://www.ruby-forum.com/.
>
> > Looking at the description athttp://www.levenshtein.net/it appears
> > that both m and n should be +1
> > The strings are then offset to allow indexing from 1.  Guess you
> > either need to add a character (blank?)
> > to the beginning of both strings, or adjust your indexes into s and t
> > (the 1st char is the 0th index)
>
> > I'm totally pulling this out of my head, as I don't have Ruby on my
> > work computer
>
> Thank you for the reply.  Unfortunately, I get the same error when I try
> this :(
> --
> Posted viahttp://www.ruby-forum.com/.

This was bugging me so I installed Ruby and gave it a go.
You do need to change that 'n-1' to be 'n+1'

Then in the nested loop you need to use both indexes:
  for j in 1..n
    for i in 1..m
      if(s[i].eql?(t[i])) then
                     ^this 'i' should be 'j'

But you still need to adjust the indexes into the strings (i-1 and j-1
seems to work here)

cheers