------art_15402_16974561.1201100709613
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

On Jan 23, 2008 7:39 PM, Alex Shulgin <alex.shulgin / gmail.com> wrote:

> On Jan 22, 10:51 am, Jan <jan.h.... / gmail.com> wrote:
> > [Note:  parts of this message were removed to make it a legal post.]
> >
> > Hi Alex,
> >
> > Have the same problem with you :)
> >
> > I think we should understand the 'adjacent' as 'not overlapped adjacent'
> > here. Each time we take two adjacent suffix, we check whether they
> > overlapped first - if so, make s2 point to next suffix, until we found a
> not
> > overlapped one.
>
> Hm... but every pair of suffixes _are_ overlapped just 'cause all of
> them last to the end of the text, so we cannot 'check first'.
> Anyway...


Yes u'r right ... I should have described more clear: take two adjacent,
check whether the "head" of them overlapped:

overlap?(s1[0, current_longest_substr.length], s2[0,
current_longest_substr.length])

Thanks
Jan


>
> > For aaaa: on the step of comparing aa and aaa, we found they overlapped,
> so
> > we keep s1  a, but make s2  aaa, the one next to aaa.
>
> This might be an appropriate hack--I'll try it when I have a
> minute. :)
>
> --
> Cheers,
> Alex
>
>


-- 
Fly on wheels.

------art_15402_16974561.1201100709613--