DSXchange

@INROWNUM only ever counts the rows coming in from the stream input link. There are no system variables for counting the rows on reference input links. It could be done with a stage variable, for example svRowsL2.
Initial value 0
Derivation svRowsL2 + Not(L2.NOTFOUND)

Beware of the Folder stage. Even though you specify just file names, it seems to try to read the entire file. This can cause an abort with access violation/SIGBUS, when your memory is exhausted by trying to load a large file for no good reason. As well as using a job sequence, you can also code it y...

Just do the other lookups using another Lookup stage.

Lance, a vector is a close analogy to a multi-valued field. A vector of subrecords is a close analogy to an associated set of multi-valued fields. Does that help any? For COBOL folks, a fixed length vector corresponds roughly to an OCCURS clause, and a variable length vector corresponds roughly to a...

Hence my opening "If you can figure out some way of loading everything after field 1 into a variable-length vector". I won't be able to devote any time to it; my current gig is server-only (so I don't have PX to play with).

If the lookup field is nullable AND the output field is nullable (and the rule is "Continue"), then you will get null. You will need to handle this null further downstream in your job.

DataStage parallel jobs use an architecture that uses an orchestra as its metaphore. There is one "conductor" process, which manages overall co-ordination and writes to the job log. On each processing node there is a "section leader" process, which manages communication upwards and downwards, and co...

Welcome aboard! :D Bienvenue! It appears that something odd has occurred during the installation, whereby the operator associated with the Transformer stage (that is, the APT_TransformOperator) has been registered twice. I suspect that the only cure will be to uninstall and re-install, but check wit...

It doesn't matter where the environment variable is set, so long as the DataStage processes inherit its value. The DataStage processes, in turn, become clients of the DB2 database server.

Beware that Char(128) is reserved for the internal representation of NULL in DataStage. And, yes, if you are getting values over 255 then you have NLS enabled, and what you are seeing is not ASCII, but Unicode code points. However these are not possible with ASCII7. Therefore, whatever "they" are pr...

Search the forum for topics like Type 2 and SCD ; this has been covered much in the past. Are you using surrogate keys? If so, search for that topic too; it changes the way you might go about designing. Comparisons can be achieved using comparison operators (such as "=") or by using the Compare() fu...

There is no specific stage type. You can use the ODBC Enterprise stage (which is probably "best"). You can also use the Dynamic RDBMS stage (DRS) with protocol set to ODBC. In either case, you will need to set up a "DSN" entry in $DSHOME/.odbc.ini describing where the SQL Server database is, and an ...

Welcome aboard! :D

Does DataStage have NLS enabled? That character set GB2312 (which is one of the Chinese maps) is specified leads one to believe so.

Do you have the DB2CODEPAGE environment variable also set to GB2312?

That, and the absolute requirement to have a space either side of the "=" character, are well-known annoyances. (It's not only in DataStage, however; for example most database's bulk loader control scripts must also have that final line-terminator, as an indicator that a command is ended.) Well done...

UltraEdit allows you to specify a fixed record length format. This might produce odd results in the first and last lines, but it's quite robust. From memory you can even skip the first line (header record).