DSXchange

Wasn't so much a question as it was a hint, I would think.

OK, so the 'distinct' part could be done with a sql query as well.

OK, so only one distinct value in column 2 for a value in column 1. Sounds like another 'fork join' scenario where you capture the number of distinct values in column2 per column1 and then decide what to do (i.e. where to go) when that count is 1 or >1.

What is your source?

Hmmm... you show "Output1", does that mean there is also an "Output2" in the picture? Do these need to go to one destination and everything else to another or are all you are interested in are the multiples? Also, why is "1 B" in your output rather than "2 A"?...

Welcome. I don't see this happening in a single job. First though, can you clarify what kind of queries are in this table - something more than simply "select * from X" I assume, yes? Are you familiar at all with concept of Runtime Column Propagation (RCP) in DataStage? I'm assuming you'd ...

Can you explain what exactly that means? Changed it from what to what?

PaulVL wrote:Since you are using a Parallel job and not one of those pesky server jobs... You can drop a peek stage into your job.

Sure, understood that. Glad we could help.

ps. Date, Time and Timestamp formats are listed here (copy the whole string, the forum software doesn't like the URL for some reason): http://www-01.ibm.com/support/knowledgecenter/#!/SSZJPZ_8.5.0/com.ibm.swg.im.iis.ft.usage.doc/topics/dateandtimeformats.html and in an appendix in your Developer's G...

sam334 wrote:So, M/DD/YYYY for January to September and then MM/DD/YYYY for Oct to Dec.

No, actually that's just for former. It means it can have single digit months (1/2/3 rather than 01/02/03) not that they all are somehow.

Time to get you into your own post... and here we are.

Next question is always going to be - what changed?

This should happen automagically in PX if the stage only includes the actual metadata for the detail records. Then anything that doesn't match (i.e. everything else) should reject.

Why don't you let us know what problem you are trying to solve, why you think you need this. Then perhaps we can provide a more proper Parallel solution.

Hello? Hello? Operator, we've been cut off!

Have you looked up the error? Talked to your DBA? Do you have indexes over the columns in your where clause? Since most people here (myself included) have zero Teradata experience your odd looking table structure or the fact that it is an 'aggregation table' don't really mean much. Hopefully one of ...