DSXchange

Use stage variables to "remember" the previous value (that is, the value in the previous row). What does "previous" actually mean in your context?

Don't try to use array binding with mixed update and insert statements. Set array size to 1.

Pivot stage pivots. There are no derivations. Any derivations, including parsing, need to be done in an upstream Transformer stage.

Welcome aboard.

You need to post questions about parallel jobs in the parallel jobs (Enterprise Edition) forum.

Persistent means "stored on disk", to contrast with a virtual Data Set which is in memory.

Can you please stop asking the same question over and over?

1996 will always be displayed as 1996. APT_DATE_CENTURY_BREAK_YEAR is defined (in DataStage Parallel Job Advanced Developer's Guide page 6-19 for version 7.5) as "Four digit year which marks the century two-digit dates belong to. It is set to 1900 by default." That is, if you have a two-digit year, ...

Yes, as noted in my earlier post.

TiimestampToDate() function with appropriate timestamp format string in a Transformer stage. date_from_timestamp() function with appropriate timestamp format string in a Modify stage. Dates themselves do not have a format, so the format only becomes pertinent if you then want to convert the date dat...

Has your site modified any of the xMEMOFF settings in uvconfig (or, for that matter, any of the settings in uvconfig file)? If so, was the regen operation successful?

Posting the same question twice won't get any better answers.
The only possible stage is Transformer.
Modify does not permit multiple argument functions (such as concatenate) while generators do not use input columns at all.

Data type string is the only possibility. You need to convert each of the non-string data types to string and concatenate.

This seems to be a strange requirement.

Why do you "have to use a Funnel stage" - is this some kind of examination? The Funnel stage requires IDENTICAL metadata on all of its inputs, so you would have to construct these based on your five different sets of metadata. And what fields are required in the output, and from which input file(s) ...

Number of lines in the file = 2 (header and footer only). This is your "abort" criterion (shudder!). Why not check before doing any of the ETL?

View Data on which link?

Please post the entire and exact error message.

Use exactly the same technique. A Cycle with a single value of 0.00.