Re: Multiple datasets in one dataset

From: Brian Tkatch <N/A>
Date: Mon, 15 Oct 2007 10:10:04 -0400
Message-ID: <3dr6h3lfc5ikngs2kaavroeqoqr5caf15t@4ax.com>

On Mon, 15 Oct 2007 02:26:04 -0700, roman.morokutti_at_googlemail.com wrote:

<snip>
>So, this would be a process flow where everything went
>fine. I could make a query where I put all the lines into
>one. But there are times when a piece crashes inside the
>machine. Thus I am never sure how many process stages have
>been passed and how many datasets are written though.
>
>The problem is to define a query where I get all in one
>line, regardless of how many stages were passed respectively
>how many lines were written.
>
>Regards
>Roman

OK, now it is beginning to make sense. Let me see if i understand this.

CREATE TABLE FOO

   (	"WP" VARCHAR2(10 BYTE),
	"VALUE_GROUP" NUMBER(4,0),
	"CURR_MS" NUMBER(38,0),
	"VALUE_1" NUMBER(13,3),
	"VALUE_2" NUMBER(13,3),
	"VALUE_3" NUMBER(13,3),
	"VALUE_4" NUMBER(13,3),
	"VALUE_5" NUMBER(13,3),
	"VALUE_6" NUMBER(13,3),
	"VALUE_7" NUMBER(13,3),
	"VALUE_8" NUMBER(13,3),
	"VALUE_9" NUMBER(13,3),
	"VALUE_10" NUMBER(13,3)

   );

1. TABLE Foo records the progress of a piece through a machine.
2. It has up to six stages.
3. Each stage has 10 values.
4. A unique piece is identified by WP and Curr_MS.
5. Different stages are identified by Value_Group.
6. The requirement is to get the current stages of a given piece.
7. An INNER JOIN query was written to retrieve the data in one record.
8. The query only returned data for pieces that completed all stages.

If that is the case, the solution would be to change the INNER JOIN to an OUTER JOIN. That would not restrict the data returned to having all stages with data.

B. Received on Mon Oct 15 2007 - 09:10:04 CDT