Matt
>I'd like to store paths to specific destinations...
See
-- Tropashko's 'materialized modell' eg
http://www.dbazine.com/oracle/or-articles/tropashko4
-- the airports example at
http://www.artfulsoftware.com/mysqlbook/sampler/mysqled1ch20.html
PB
-----
Matt Juszczak wrote:
> Hi all,
>
> I've got a table such as the following:
>
> id1 char
> id2 char
>
> sample data looks like this:
>
> id1 id2
> 1 3
> 2 4
> 3 5
> 5 6
> 6 8
>
> And of course another table has something like:
>
> id info1 info2 info3
> 1 blah blah blah
> 2 blah blah blah
>
> I'd like to store paths to specific destinations...
>
> In other words, the path from 1 to 8 is:
>
> 1,3,5,6,8
>
> I was thinking of creating a table called relationships
>
> start end path
> 1 8 {3,5,6}
>
> This would allow me to easily display the path if I know the start and
> end, but what it doesn't allow me to do is reuse the data.
>
> IE: say that I calculate the path from 1 to 8 as 1,3,5,6,8, and then I
> want to know the path from 3 to 6. even though this is already
> calculated, I have to recalculate it as another row... hence
>
> start end path
> 1 8 {3,5,6}
> 3 6 {5}
>
> I considered making another table, called hops, such as:
>
> start end relationshipID
> 1 8 1
>
>
> table hops:
> relationshipID start end
> 1 1 3
> 1 3 5
> 1 5 6
> 1 6 8
>
> Then I could almost "reuse" those hops somehow.... but not sure.
>
> Can anyone recommend a good way to store this data?
>
> Thanks!
>
> -Matt
>
