Re: Nearest Common Ancestor Report (XDb1's $1000 Challenge)

From: Neo <neo55592_at_hotmail.com>
Date: 1 Jun 2004 18:08:03 -0700
Message-ID: <4b45d3ad.0406011708.632c1618_at_posting.google.com>


> I changed my model ... now down to 11.0 ms...
> better than the 16 ms that XDb1 needs ...

The above are apple-to-orange comparison for many reasons. This might be one:

To test out the genericness of the solutions (especially the db data/schema), I added the following small hierarchy to the existing ones in the db via XDb1 (v4.5.9):

mary parent john.      (equivalent to 'mary isChildOf john.')
luke parent john.      (equivalent to 'luke isChildOf john.')

Right-clicking john and running report produces:

Common 'parent' Report for 'john'

ThingX	ThingY	CmnAnc	Dist
mary	luke	john	2

Time Elapsed: 1.029314 msec

Note: Because the older version has problems passing report parameters, it won't work, unless extra steps are carried out via the GUI. Requires moving john's relationship to luke before other hierarchal relationships.


In RM Sol#1, I added the equivalent data:

insert hierarchies (thing, otherthing, hierarchy)

       values ('mary', 'john', 'parent') insert hierarchies (thing, otherthing, hierarchy)

       values ('luke', 'john', 'parent')

The command "exec CommonAncestors _at_hierarchy='parent'" produced the following empty report:

Starting time 2004-06-01 19:30:35.390
Ending time 2004-06-01 19:30:35.420



I am assuming the report is the same in sol#2. Is this due to db's data/schema and/or report procedure? Could you update it/them to make it work? Then we would be one step closer toward making an apple-to-apple comparison. Received on Wed Jun 02 2004 - 03:08:03 CEST

Original text of this message