On Mon, Sep 18, 2006 at 11:12:54AM +0200, Gevik Babakhani wrote:
Magnitude comparison on the GUID as a whole makes no sense to me.
I agree. Any kind of comparison except equality has no meaning for the
GUID. (And this is discussed before) I rather have the option to sort
and group for the sake of consistency and compatibility.

Thomas: The ability to sort / comparison is required for use with
B-Tree index. I prefer a fast comparison over one with more meaning.
memcmp() is fine with me and it is how I implement it in my UUID
PostgreSQL library.

Fair enough. Although the magnitudes as such makes little sense, the ability to order will make it possible to compare results from different queries etc. Very difficult to do with random order.

Thomas Hallgren

---------------------------(end of broadcast)---------------------------
TIP 6: explain analyze is your friend

Reply via email to