DSXchange

I am trying to run the grep command in an external filter stage. But I get an error saying

Field X is null but no null export handling is defined

Didn't know we had to handle nulls here too. How do I do this?

The number of files will vary everyday. We are performing validations on a per-file basis based on the suffix of the file.

Hi, I am posed with an elementary level issue. The requirement is to read multiple files with similar schemas and load into a table. The problem is that each of these files have column names as header. When I try reading all the files of a specific pattern using the 'FILE PATTERN' property in the Se...

That should work too and looks a lot easier on the eye

But I hope I'm right about the Octal part. Would like your feedback, Ray

Besides, I think he isn't trying to implement an Oracle function. Rather he's trying to implement the DataStage equivalent of the Oracle Decode function.

I'm not saying 09 is octal. All I'm saying is that 8 and 9 do not appear in the octal system. When we say somethinf like x=08, DataStage seems to look for an octal value. If however, the values are enclosed in quotes or instead of x=08, we use x=8, the problem seems to get solved. I don't know if th...

Try this in_Xfm_2.MO_NBR= "08" or in_Xfm_2.MO_NBR= "09" Looks like DataStage is looking for Octal values when you try equating a number padded with a '0' before. Put the non-octal values within double quotes and I'm sure your problem will be solved. Else IF you can, then just wri...

Just FYI.. The CHAR function in DataStage returns an ASCII character from its numeric code value. Is this what you want to do?

What if MO_NBR is null? Shouldn't you be handling the 'null' scenario before implementing your logic?
Implement a null handling logic before this and you should be able to get through successfully.

Also, are you sure FYEAR is an integer /number ?

Just a hunch. I think his target is a Timestamp field rather than a 'String' that holds a timestamp value. This could be the reason why DataStage is trying to compare the output of the TimestampToString with the default DataStage format and consequently warning the user about the mismatch

I believe that this is the default behaviour and you won't need any dedicated logic to accomplish target nullability. If all the values of your aggregation column come as nulls, then you can either choose the output sum to be output as 0 or NULL (see 'Allow Null Output' option in Aggregator). This w...

Use TimestampToString(CurrentTimestamp(), "%mm/%dd/%yyyy %hh:%nn:%ss")

I am relatively new to this forum and am still getting used to the 'Best Practices'. Your information helped that cause. Thank you

Never mind. Solved this by setting reject failure to 'Continue' and filtering the resulting nulls.

Just one more scenario that I am posed with. None of my keys are nullable. Basically all the reference links carry an ID for a corresponding key in the stream link. Sample Record in stream: A,B,C,D,E There would be an ID in the reference link corresponding to each column s that my target would be A,...