DSXchange

Change your routine so that the output and exit status are logged. $DEFINE DEBUGGING $IFDEF DEBUGGING Message = "Output from command " : Quote(Arg1) Message<-1> = "Exit status = " : SystemErrorcode Message<-1> = Quote(Convert(@FM,"~",Output)) Cal...

There are various techniques available. Search the forum for "vertical pivot".

Hi all, like me new startes the question and answers your posted is like greek and latin. most of the answers in dsx are wierd and we are not getting any sort of help from this forum. please explain the basics concepts and get into deep and we can activily particapate and share our knowledge and i ...

Have you tried to declare a decimal(0) or string[0] or similar in the fifth field?

The best way is to do it at source, when loading the Data Sets.

Otherwise you need to introduce a prior task (a server job?) to collect the current max value, and supply that as a parameter to the job that is splitting the Data Set.

Welcome aboard. :D

If you test for validity before applying the conversion function, and only apply the conversion function to valid dates, then the warnings will not appear.

You have to specify, and then design, what happens to rows in which the date is not a valid format.

No, as you have already established in a separate thread.

Why not do your own research? Start here for example.

If you don't want to, I would be happy to do so for a suitable (to me) fee.

Just right click on the warning itself and create the message handler from that menu.

The "Error Codes and Actions" manual is promised for the next version (Hawk). Until then, there is no single source of message codes (at least as far as I am aware) in the public domain.

Three job designs. Two as you state, and a job sequence to control whether or not the second job (the load job) executes.

You don't.

That would not make sense, and would slow down execution to an unacceptable degree.

You do it in an after-stage subroutine, an after-job subroutine, or in a routine invoked from a Routine activity after the job has completed.

It's not the only option. As tables become larger they can become slower to load. You can prove this by replacing your Oracle stage with a Sequential File stage and watch the speed return to fast. Have "they" put additional indexes or constraints on the table? That will slow you down too. Once you h...

Search the forum for SPINTRIES and SPINSLEEP

Supplied functions (those in on-line help and manuals) don't need a prefix.