|
Posted by bevanward on 05/07/06 01:38
Erland really appreciate the effort you've put into this message it has
helped alot.
The reason there is a key as well as the location data is that is the
number the samples are collected and tacked against. There can be
points with the same coordinates in real life however there is a
problem where the coordinate location is confirmed with a more precise
method and it is given a new name - ideally it would not be possible to
enter this into the database however since the planning phase is not
mandatory it has to have a merge character... I hope in the future the
work flow can develop to stop these types of things happening.
I wasn't able to get the query to use the index (I do have a clustered
index on x, y, z) with either the cross join restricting on the where
or the full outer restricted on the null key.
I've had to give up on it for now but have printed your message and
will try to get something effective working - I will test a cursor
later today and see how it will run.
Unfortunately this database is still in 2000 however once it moves to
2005 I can see as you state this problem can be solved in a more
conventional manner.
Thanks again for your effort and I hope to be able to post a method
back that runs more efficiently in the near futuer.
Cheers
Bevan
Navigation:
[Reply to this message]
|