DSXchange

With server jobs, you can "display performance statistics" in the job design and watch what's happening - provided you don't mind the impact on throughput that your interference causes! Some form of visual indication in a Job Sequence of progress would be really nice to have. For example, highlighti...

Something like this ought to do it. FUNCTION FindCode(SearchString, CodesList) * Initialize return variable. Ans = "" * Convert CodesList to dynamic array for efficient search. CodesArray = Convert("*", @FM, CodesList) * Search until either SearchString is found o...

Matthieu, are you talking about parallel jobs? It's my understanding that - unless you're using the "BASIC Transformer" stage - implicit data type conversions do not occur; this is a behaviour found in server jobs only.

What does the routine need to return? Only the first match, or all matches, on the first two characters? How are you specifying the search string (for example "CD")? Is it an argument to the routine? How are you specifying the codes string? Is it an argument to the routine? The solution will probabl...

What did you expect to get? Examine your expression from the inside outwards. You have "1234.567" in a VarChar(8) as source. The first thing you do is to remove the ".". You would now have "1234567". The next thing you try to do is divide "1234567" (a character string) by 1000 (an integer). Is this ...

Where is fastload.exe?

Is that folder mentioned in the search path environment variable (PATH)?

Follow the link in the third post on this thread.

DataStage logs a warning whenever there is unexpected output, on the basis that, if there is unexpected output, you ought to be alerted to take a look at it. Obviously a PRINT statement will generate output that DataStage does not expect. The algorithm is not rocket science. If the output file from ...

Welcome aboard! :D Refer to Chapter 20 of the Parallel Job Developer's Guide for full information on using the Lookup stage. You're on the right track, you identify the lookup keys (and whether case sensitivity is required and whether you want to allow multiple return rows) in the Input link propert...

PilotBaha beat me to it. Yes, it's a simple as creating a redefinition of the input file, so that there's another name that refers to the first nine characters of CARDHID. You can then perform any kind of match between this and the nine character column in File B. PS You can still learn COBOL, but i...

I would be very wary of anything (such as funding from Ascential) that would constrain the independence of DSX. Sure, sell them advertising space, but never control over content. Sponsored links, like those at Google, might raise some funds too, maybe from headhunters and the like. The donations box...

The vendor is MainSoft (formerly known as Data Direct - I do not have current pricing information, but can inform you that it's per CPU in your machine, and expect tens of thousands of US dollars. Search the Forum for examples of code using BCI functions - there are quite a few that have been posted...

"Job not in a runnable state" means that the job needs to be compiled or reset - either it has never been compiled, or it's most recent run was stopped or aborted.
Did you compile your job sequence?

If the DataStage metadata specifies CHAR and your SQL specifies CAST(column AS CHAR) then you will not get a metadata mismatch. The metadata are compared against the result set on a SELECT, not on the base tables.

You may have extra line terminators in ScreenOutput. These will have been translated into field marks (@FM). Try: