Re: SQL challenge

On Tue, 13 Jul 2004 11:48:21 -0700, Mikito Harakiri wrote:
> "Jan Hidders" <jan.hidders_at_REMOVETHIS.pandora.be> wrote in message
> news:pan.2004.07.13.17.45.22.985796_at_REMOVETHIS.pandora.be...

>> Well, the problem was to find the *minimal* set. [...]
>>
>> You're not there yet. :-)

Yes, you could approach it that way, but there is a simpeler way.

> Another glitch is for each triangular face (p1,p2,p3) the output would have
> them in all permutations
> (p1,p2,p3)
> (p2,p1,p3)
> (p1,p3,p1)
> ...
> so that I have to take care of this too.

That's easily solved by for example only selecting edges where the first point is lexicographically smaller than the second point.

> Solving these small problems and integrating them into the answer doesn't
> look particularly challenging to me:-) but, i'm afraid, it might lengthen
> the final solution a bit.
>
> Is there an elegant SQL statement? If you say yes, then, I would think hard
> finding it:-)

Yes, there is. We (some people in my research group) found it ourselves, but we had to make the assumption that not all points are in a single plane. Without that assumption things seem to get ugly.

> BTW, after posting the first attempt, I gooled Jims 2002 paper. He seems to
> consider planar case only, right?

Correct. But we already figured out at the conference how to do this, so that was not much of a puzzle.

• Jan Hidders