DSXchange

Hi,

OR logic is not working as expected in the transformer and hence recently even we had to change all the OR conditions to If Then Else in the transformers. I think Sushant is also a victim of this issue.

Thanks

Bob you are exactly right. I want F3 = F1 but without using a transformer stage. I tried using a copy stage to map F1 into F3 but got "Invalid derivation for output column F1." compilation error in return. So I was trying to use a modify stage. Now the modify stage has 2 input columns and ...

Hi Rich, Let me clear the confusion. Field F3 should be NULL if field F1 is null else it should be equal to field F1. Hope this clears the confusion. Also note that field F3 is not present in the input of the modify stage but I want it to be present in the output of it. Can this be done? Your answer...

Hi All, I need some help on the usage of modify stage. Let me first explain the scenario. We have two fields in the input say F1 and F2. We want a third field F3 in the output(in addition to F1 and F2) whose value is decided as follows : F3 = F1 if F1 is not null and F3 = NULL if F1 is null. The abo...

Hi Gurus, I am facing problems while viewing data from a fileset when I set the format as Fixed-width records on the Fileset stage. The meta data definition for the file contains fields with datatype BigInt, SmallInt, Integer and Char. There are no VarChar fields. I have also specified the Record le...

If i understand Inquisitive correctly, you want to split a DataSet just like a flat file on Unix. No, you cannot do this. Just because the data in a Dataset is already split and stored in nodes. If you want to distribute your data into various datasets, do it in the job itself, by having multiple ou...

Is the precision of the summation wrong or the value itself is wrong??
Can you eloborate more, or give more inputs.

Regards,
Sandyla.

Its true that, we cannot capture the unmatched records in a JOIN stage, but we can always do that by using a Transformer stage following the JOIN stage. IsNull(leftKey) - by using this we can figure out if there was an unmatched record in the left link. leftKey="" - should work if its a VA...

Hi, We are facing a severe issue in our parallel jobs. My jobs after processing the input data are writing it into a dataset. The table definition(order of columns) of the dataset and the mappings are in the correct order. The issue is, when we make some minor changes to the code (not in the mapping...