DSXchange

The answer to the question is "the same NLS settings that the data is encoded in". What are your ORACLE NLS environment variable settings and have you checked your oracle table nls settings?

did you include "math.h"? Are the math libraries in the search path at compile time?

The Shift-In and Shift-Out characters will be the same only if you ensure that you define the two stages (CFF and database) with the same NLS settings, otherwise you won't be able to compare the two streams.

yimwai - that's a simple and elegant solution

You can use job parameters to pass in dynamic query statements for the ODBC stage.

It would seem not, I just played around with the aggregator stage and couldn't find a way of doing with just one stage. Perhaps someone else has found another solution.

Will the INT() function suffice for your purposes?

Try

Once place that the project's status as protected or not is visible is in the UV.ACCOUNTS file, the following query should list all protected projects: LIST UV.ACCOUNT ID.SUP @ID WITH EVAL 'FIELD(NAME,"\",4)' = 46256 A BASIC routine to return 0 for unprotected and 1 for protected would be ...

As has been stated above, you are getting NULLs in your target because the converion from string to numeric in DataStage is failing on certain values. Do an explicit conversion in DataStage and you'll find out which values are causing you problems.

What happens when you remove the "int ParsenExtract(char *,char *,char *); " from your source code?

So the source data looks like "2010-01-01 12:30:44" and you are reading it into your job as a CHAR(19) or VarChar(19) or VarChar(), correct?

Could you post the exact error messages for (2) and for (3)? Particulary the sequential file issue is strange since DataStage doesn't really "connect".

The warnings about conversion from single to double can be avoided if your SQL select does a "CAST(YourColumn AS DOUBLE)", but the question is whether or not it results in the value you wish? Also, in the original job, could you try to keep the data type, but put in a transform derivation ...

can you do a tnsping to your database from the DataStage server?