DSXchange

You're doing something fundamentally wrong. The expressions you use in a Transformer stage are exactly that; expressions. They are not assignment statements. To set a variable value you must do it indirectly, perhaps in a Routine that copies the value into the variable then returns the value unchang...

Create two jobs, run them with dependency in a sequence. The first job contains nothing but execution of your script (in Job Control routine). Or in a before-job subroutine, but you will then need to do something innocuous, such as X = 1 as your "job control" routine. The sequence will detect the co...

Yes, as far as I can remember, but you have to be VERY careful to use the correct version of the Informix client software and/or ODBC driver.

How about this then? You don't need the Convert statements if you can deliver your values as dynamic arrays - I have used comma delimited lists as examples. Every function I've used is in the DataStage BASIC manual. StartVals = "1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20" EndVals =...

Ignoring DataStage, what are you trying to accomplish? When we know this we may be able to supply more cogent answers. There are many ways to execute a command from a server job. Probably the easiest is a before/after subroutine, which is executed only once. I take it you don't want to execute the c...

Stage variables will not have been evaluated when a before-stage subroutine is executed, so would generate "variable not assigned" errors if you could do it. The DSGetStageInfo() function will allow you to retrieve a list of stage variable names . However, when the Transformer stage is compiled, the...

If the meaning of "duplicate" means "same key", write to hashed file then read from hashed file. Updates to hashed files are destructive overwrites, so this is an easy way to remove duplicates. If you need to record duplicates, this is slightly more complex, in that you have to check the same hashed...

The shmtest utility does not make assumptions about huge hashed files.

Indeed, its main purpose is to detect whether the default locations are in use by anything else. I understand (based on only a small amount of investigation) that it recommends the default locations if they're free.

Umm... What do you mean by "dataset" here? In the PX environment we usually refer to a dataset as something created by DataStage PX. Also, a fileset is something similar, the main difference being that a dataset is in memory to the greatest degree possible, while a fileset exists in persistent stora...

It always has been and, I believe, always will be, DataStage philosophy to log every item of information received from database servers, even though these are only informational messages. In the highly litigious USA, this limits the possibility of Ascential being sued because their software failed t...

Beware that all DataStage function names are case sensitive. Further, the link row count is obtained from DSGetLinkInfo, not from DSGetJobInfo function. As a graphical alternative, you could generate two more outputs, with a counter (fed by @OUTROWNUM), pass these through Aggregator stages applying ...

WoMaWil wrote:In Ray you can trust.

Luckily I know Wolfgang isn't in the USA, otherwise I'd be forced to comment that this is a bit unnecessarily messianic!

In general no. If you're only reading from the hashed file, there is no contention or locking. Though if you're looking to load the hashed file into memory, read the manual about sharing cached hashed files. If you're writing using multiple links, you need to be aware that every write to a hashed fi...

Write that link to a sequential file, then read it again with tail -1 as the filter command.

Poor skinless cat!!