DSXchange

Stored procedures called from within DataStage jobs must process at least one row of data. "Stand alone" suggests that your stored procedure does not fall into that category. If you have a licensed ODBC driver you might contemplate using a routine using BCI functions; otherwise you will need to use ...

There is no way to do this using a Hashed File stage which, conceptually at least, performs an equi-join on the key column values. Given that the data are in a hashed file, you can use a UV stage to access it using conventional SQL, which allows for a BETWEEN selection. Use the pre-defined DSN local...

I don't know how to do it from the UNIX command line. Even if you could do it from the DataStage command prompt (>), there are way too many separate components to drop or remove (as appropriate) ever to be conveniently loaded into a single command. To create such a command (set of commands run from ...

Just to set the record straight, I am not, and never have been, an employee of Ascential. On this basis it is unreasonable to expect that I am privy to their plans, other than those that they publish ("roadmaps"). All of you have access to these; I take the trouble to study them. I did work for VMAR...

How many rows per second does sqlplus deliver? The problem could be in the network connection itself. Are you using identical SQL statements in your OCI stage and in sqlplus? (If not, your comparison is not valid.) Remember, too, that rows/second is a very misleading figure. The seconds include all ...

Guess what the defining expression for the DIGITS Transform is?

Don't you thing that DIGITS is a more easily remembered name than Oconv(Arg1, "MCN") ?

The end of file character (Ctrl-D in UNIX, Ctrl-Z in Windows).

It's probably impossible to check for when using the Folder stage. Alter your design so that empty lines are filtered out (for example, filter through grep -v '' in UNIX)

There are two criteria: number of rows and maintenance. Number of Rows Use a static hashed file if the total number of rows is known in advance and unlikely to change very much. Use a dynamic hashed file if the number of rows is not known in advance or it is known that the number will vary markedly ...

Why are you using a routine at all?!! A lookup to the hashed file would serve as well. The same hashed file can be kept up to date within the job by writing to a second Hashed File stage, remembering to set lock for update in the stage that reads from the hashed file. This technique is taught in the...

Read the on-line help about DSAttachJob(). Its second argument determines whether, if the controlled job aborts: (a) the controller aborts (b) the controller logs a warning but keeps going (c) the controller keeps going without logging anything Automatically-generated job control code chooses option...

That's a question that rightly should be directed at the vendor (Ascential Software). If it can not form aggregate functions properly on character data, even though the data contain only numeric characters, this is a bug.

This directory is used when, in Designer, you select "create job from template" or "create template from job". The ones provided are, or support, the "intelligent assistants" you would have read about in the readme files. Not much more to say, really... Except that they are local to one PC. However,...

This is no consolation, but I recently encountered the same thing on a 7.1 site. I think it's a (re-introduced?) bug.

Luckily I was able to what I needed to do using stage tracing.

The solution will most likely involve stage variables, and require that your input be sorted. Read about stage variables in the DataStage manuals. Alternate solutions exist, involving Routines and variables declared to be in a COMMON area in memory. The RowProc category of SDK routines make use of t...

If you would use the Expression Editor to create this expression you would discover that input column names must be qualified with input link names. String constants must be quoted, as in most (all?) programming languages. Hence for example, unless it's the name of a stage variable, D2D is meaningle...