Re: Encoding materialized path in an atomic value.

Mikito Harakiri wrote:
> Vadim Tropashko wrote:
> > Encode all the numbers in unary numerical system (written backwards for
> > convenience), e.g.
> > 1 - 1
> > 2 - 01
> > 3 - 001
> > 4 - 0001
> > Encode each number in materialized path with these, e.g.
> > 1.5.3.2
> > 1000100101
> ---^^^
> 1000*0*100101
>
> This is quite an interesting perspective. Given that boolean vector is
> a set,e.g.
>
> 10000100101={0,5,8,10}
>
> then, we have node A being an parent of node B whenever set A is
> containted in B. Pardon, but that's exactly nested sets (with
> containment operation reversed to celebrated Celko's Nested Sets)!

No, prefix string containment is not the same as set containment. Consider

1.1.1 = 111
1.2 = 101

111 and 101 are incompatible as strings, yet 101 is subset of 111. Received on Sat Sep 24 2005 - 03:22:07 CEST