DSXchange

Given that we're in the server edition forum, I'd say Craig has nailed it. That form of date can be converted with the "J" control in Iconv/Oconv. Here is an example:

It's a year since this material was first produced, and it's probably due for a review. So I'd like to invite anyone who has attended one to provide constructive feedback, covering areas like topics that were included but could be omitted, topics that were not included by should have been, and any w...

Curiously, perhaps, William of Ockham was bearded for much of his life.

The term "razor" is not the one related to shaving, of course.

kali wrote:How do you know all these things?!

It's the difference between 12 years of experience versus 12 weeks of experience! Plus the luck to have been a UniVerse specialist working for VMARK when they brought out DataStage and the good fortune to have been able to continue working with the product.

Welcome aboard. :D The short answer is "probably". Do you want to pivot (generate a line for each of a set of columns)? If so, use a Pivot stage. Do you want to return multiple rows from a lookup? If so, enable multiple row return in the Lookup stage. Otherwise, what is the basis for generating new ...

Is the deadlock daemon running? If so, give it a chance to run through another cycle or two - it may clear whatever is locking the project. It may not be one of the locks visible through list_readu - there are others. If dsdlockd is not running, then invoke it manually then check its log to see whet...

Can't help with the first question. The Help on DSPrepareJob is unhelpful in this regard, and the DSGetJobInfo help does not seem to cover the topic at all. Yes it can certainly be done from RT_STATUSxxx, but that's venturing into unchartered, not-supported-in-version-8, waters, and I'm trying to we...

I wouldn't use a Wrapped stage (note the correct name) within a job; I'd use an Execute Command activity in the controlling job sequence, much simpler. A Wrapped stage in a job must execute for each row processed. You would need to define the input and output interfaces to the Wrapped stage. You pro...

Does your table definition (Columns tab) include a VarChar or Char with nothing in the Precision column? There's no such thing as a VarChar(0).

If you're uncomfortable with WRITESEQ and the BASIC family of sequential file handling statements, an alternative is to use an echo command. Shell = "UNIX" Cmd = "echo " : FinalMsg : " >> /u04/test/FinalFile'" Call DSExecute(Shell, Cmd, Output, ExitStatus)

Start by posting the exact error message. Improper data type may be completely unrelated to your arithmetic. Also post your job design.
You might also benefit by searching the forum for the error message text.

You don't, unless you're referring to the "Don't sort, already sorted" or "Don't sort, already grouped" settings on the input link sort. Can you provide the page reference and manual name where you found this quote?

There's whatever you can find. Every site, every project, has its own Repository of re-usable components. Every developer should learn what's there, so as to be as productive as possible and not need to reinvent any wheels.

Use the Debugger or stage tracing to process a few rows and convince yourself that there are nulls appearing from somewhere (a failed lookup, for example, will generate nulls in the same way that a left outer join will). Then set up some stage variables so you can handle the various scenarios withou...

You may (will) also need to assign values to your SYBASE environment variables in the DataStage dsenv script.