DSXchange

N rows in ---> Stage ---> Stage ---> Stage ---> (<= N) rows out Why do you think your design would generate extra rows? Adding extra Transformer stages would not generate extra rows (though can constrain some rows from passing. Aggregator stages also typically reduce the total number of row...

No.

You need a different tool for that. It's called a DBA.

The Selection tab is never enabled if you are using the hashed file on a reference input link. Being local or remote is irrelevant. There is no reason that the Selection tab would be disabled on a stream input just because the hashed file is not in an account. At my current site we have the same rul...

The different instances you see in Director are like views. All instances share the same physical log (and there is another reason for creating jobs that don't generate millions of warnings!); that the job was run with an invocation ID means that there is a separate log view for that instance. Recom...

You can use the three lots of identical column definitions in an OCI stage, or a DRS, or any other stage that accesses the database. Here's how. Load qualified table definition for the first table. If the column names are not qualified in the Repository, edit the Derivations column and make them qua...

Then there's the obvious question, why are you "writing an Aggregator in a Transformer"? Why can't you use the Aggregator stage that's part of the product?
The Aggregator stage in PX is sometimes unfairly tainted with the bad press that it's server job equivalent has received.

Then you'll want to be able to do more than one! How many databases allow multiple SELECT DISTINCT in the one query? For example SELECT DISTINCT COLUMN_01, DISTINCT COLUMN_02 FROM TABLE; or SELECT COUNT(DISTINCT COLUMN_01), COUNT(DISTINCT COLUMN_02) FROM TABLENAME; Not many. (Red Bri...

No, mkdbfile does create the dictionary component, and SETFILE includes a pointer to it. You can prove this with a command such as LIST.DICT hashedfilename The thing that doesn't happen when you use mkdbfile is the creation of column definitions within the file dictionary (after all, it's got nowher...

It's highly likely that the indexes need (re)building. The command for this is DS.REINDEX ALL. You need exclusive access to the project.

You can check first: LIST.INDEX DS_JOBS ALL is a good first check.

But wait, there's more! All design objects are exactly that. When you put a design object (a stage or a link) onto a design canvas, you are instantiating an object of that class. So, when you look at the stage types division of the Repository, you see names like CTransformerStage, which means "the c...

My point exactly. If they have UniData, they have the ODBC driver for it.

Of course, theoretically, you can. Unfortunately, to do so you would require knowledge about where in the DS_JOBOBJECTS record the collection of column definitions resides, and this information is not in the public domain. And Ascential reserves the right to change this location at any time: actual ...

Whenever possible apply constraints as early as possible; in this case when selecting from source. It will help if the constrained columns are indexed. By eliminating rows as early as possible you are minimizing the total volume of data to be processed and therefore minimizing the total time for com...

DS.TOOLS has been in the product since version 1.0, and has been undocumented for all that time.

I don't believe the XMLReader or XMLWriter have any ability to acces URLs. We use them in converting XML documents that are received from and sent to .NET applications via RTI. RTIInput ---> XMLReader ---> DataStagePieces ---> XMLWriter ---> RTIOutput The DataStagePieces include some QualityStage st...