Archives of the TeradataForum

Message Posted: Thu, 21 Jan 2010 @ 10:05:22 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

<-- Anonymously Posted: Thursday, January 21, 2010 03:59 -->

Thanks Victor,

firstable, to answer your question:

     case when R.ID_MINUTE = A.MINUTE_STARTING THEN A.SEC_FIRST_MIN

should be

     case when R.ID_MINUTE = A.MINUTE_STARTING THEN 60 - A.SEC_FIRST_MIN ?

It should stay as it is, but basically you are right - I made a mistake when explaining the meaning of fields. SEC_FIRST_MIN is actually a number of seconds used within the 1st minute, so what you propose was already done when the value was inserted into the field.

My fault, sorry about that.

However, you understand the meaning of the query well.

Unfortunately, I am on V2R6, so no chance for scalar queries.

I am not sure if I can avoid joining the reference table only by extending the CASE though. I need the data to be grouped by every minute- slice, while the main table stores just the ID of a starting and an ending minute. Maybe I miss something here, but I don't think it will be so easy.

So OLAP joins the game probably, which I really wanted to avoid. Besides, I am not sure if I can use OLAP to get what I want - I'm afraid I will have to rack my brain a bit more.

	<Prev	Next>		<<First	<Prev	Next>	Last>>