Null Substitution in Joins

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
amit_dwh
Participant
Posts: 22
Joined: Tue Apr 11, 2006 6:04 am

Null Substitution in Joins

Post by amit_dwh »

Hi

I am encountering a strange thing while performing join in Datastage.

When i perform a left/right join on on two datasets in DS then it substitutes Null for non matching records but when i perform a join between a dataset and teradata enterprise stage then it substitutes :
1) Blanks for character values
2) Zeroes for integral values

for non-matching Teradata Enterprise data but NULLs for dataset non matching values.

So is it that NULLS are not substituted in DS after joins for non matching Teradata records?
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

This looks like default value substitution. First, though, can you be certain that the substitution is occurring after the Join stage, not before? If so, can you check these fields' extended properties to determine whether a Default is specified for the particular data types?

A "default default" such as you have described can also occur when NULL is generated in a stage but the target field is defined to be not nullable.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
amit_dwh
Participant
Posts: 22
Joined: Tue Apr 11, 2006 6:04 am

Post by amit_dwh »

i checked it and DS substitutes NULLs after joins only for datasets and not for databases.
Post Reply