Re: I: About "Our CLUSTER implementation is pessimal" patch

Leonardo F <m_lists@yahoo.it> wrote:

> Attached the updated patch (should solve a bug) and a script.

I reviewed your patch. It seems to be in good shape, and worked as
expected. I suppressed a compiler warning in the patch and cleaned up
whitespaces in it. Patch attached.

I think we need some documentation for the change. The only downside
of the feature is that sorted cluster requires twice disk spaces of
the target table (temp space for disk sort and the result table).
Could I ask you to write documentation about the new behavior?
Also, code comments can be improved; especially we need better
description than "copy&paste from FormIndexDatum".

Regards,
---
Takahiro Itagaki
NTT Open Source Software Center