DSXchange

Why didn't you ask in the parallel forum?

Assuming you have a string data type you would use something like
If the incoming data type is not string, you would need to convert it to string.

You need to parse the comma-delimited field into separate columns, either when reading the file or using Field() functions in a Transformer stage upstream of the Pivot stage.

You don't need any parameters in the Pivot stage. You do, however, need to have parsed the comma-delimited format into separate columns.

.

ogmios wrote:And wasn't NLS only included as of 6.x ... I don't recall 4.0 (as in the OP) having NLS at all.

The first version of DataStage to have NLS was 2.5, the next was 3.6, then every version after that (except x.0 versions) has had NLS capability.

.

ogmios wrote:And wasn't NLS only included as of 6.x ... I don't recall 4.0 (as in the OP) having NLS at all.

The first version of DataStage to have NLS was 2.5, the next was 3.6, then every version after that (except x.0 versions) has had NLS capability.

Are you running version 7.5 or something earlier? If 7.5 there is a sample routine that shows how to use DSSendMail with an attachment.

That won't work in a parallel Transformer stage; Maximum() is a DataStage function that applies to dynamic arrays. So, instead of the Maximum() function you need three logical expressions joined by OR conjunctions. That is, define svMax as (sv1 > param) Or (sv2 > param) Or (sv3 >...

You need to Search the forum to discover the answer to this question, then have your DBA grant the appropriate SELECT privileges.

Probably. If something already exists to perform the extraction, use that. Otherwise ABAP is the easiest way to do it.

OK, there is a Sort stage upstream of the Remove Duplicates stage, is there not? In this Sort stage you must sort on the keys on which you are going to base the "duplicate" decision in the Remove Duplicates stage. Data must also be partitioned on these keys if the right results are to be obtained.

You remember correctly, but there is a caveat that none of the "go forward" plans is guaranteed to happen. Yes, it's on plan for version 8.1.

1. No. 2. No. It's not possible. You can report separately on the subroutine. But not using dsjob command. You may also be able to do something through the Information Server reporting console but you won't be able to combine the job and its before/after subroutine(s) without hacking the generated X...

That's right - it needs to be executed on the same machine as the Transformer stage. That means that a VB6 DLL can't be used; it is specific to Windows. You need to create object or function library that can be called in a UNIX environment. This rules out the use of Visual Basic as the programming l...

Welcome aboard. There is a useful Search capability in the menu at the top of the screen. You might start by using it. dssearch is one of those delightfully undocumented commands that suddenly appeared in the product one day. It probably gets a mention in the documentation that comes with the softwa...

English is my first language and even I struggled to make sense of that. Please use professional standards of language on DSXchange. Have you considered doing the parse check in one job and the actual processing in another? I think you'll find such a modular approach easier to understand and to mai...