DSXchange

Rule of thumb: the "#" characters are required in passive stages but not in expression fields in active stages.

In parallel jobs, active stages are those that may have both an input and an output link.

Learn how to use DSXchange's Search facility - you could have had this answer without waiting for anyone to post a reply to your question.

John, that was never intended as a put-down, just an alternative. I like Perl but - purely because of my history - am way more proficient with DataStage BASIC. And some variant of C could as easily be used.

Good analysis of the partner property, too.

Does your job include one or more Transformer stages? What settings do you have for APT_COMPILE_OPTIONS and APT_LINKER_OPTIONS? What version of the C++ compiler are you using?

What is the target? In particular is the target a text file or something else?

No.

But you can write a script to do that and invoke the script from DataStage.

Search DSXchange for "vertical pivot" for this is what you are attempting to achieve.

No you have not got the reason, you have removed a cause. I suspect that the reason was an unquoted date.

In the DataStage BASIC manual you might also investigate the Fix() function.

No limit, apart from any imposed by the operating system. What do you mean by "performance" in this context?

Version 4 does not have job sequences. You would need to do this using a job control routine. And - I believe - you'll need to construct a script to call the stored procedure unless it deals with a record set, in which case you can use an ODBC stage.

Looks like you've bumped into a limit in the product. You might like to try using multi-word pattern elements such as ? or ** or positional or substring pattern elements.

Make sure that the keys to the hashed file and the values that you are looking up are identical - in particular identically trimmed.

Not really unless you can provide an example or two of your input, the pattern(s) generated from your input, the pattern-action set that you tried and, most important of all, the exact error that you obtained.

WHERE does the "q" appear?

What data type?