Converting from Oracle to Teradata gave me some surprises when transferring data (primary index errors, unexpected dupes). In Oracle we operate in a CASESPECIFIC mode, but I found that by default our Teradata instance is operating in Teradata mode (NOT CASESPECIFIC). We would like to create tables in CASESPECIFIC mode and treat all of our character columns as CASESPECIFIC. Is there a way to specify the transaction mode when running a TPT/tbuild script? I have checked the manual and found no mention of this.
TPT always runs in Teradata mode.
It is hardcoded that way and the user cannot change that.
dbs controll field “DefaultCaseSpec”
Determines whether character string comparisons consider character case and whether character columns are considered case specific by default in Teradata session mode.
But this is a system level field!
I would really try to avoid it - case specific fields. Often they are not needed.
Thanks for the replies. We work with medical data so there are often times where mixed case makes sense.
Since TPT can't log in with ANSI, then I guess we can set CASESPECIFIC on the individual columns in the TPT script's CREATE TABLE statement? Does that sound correct?
Go to the below link
You will find the materials for TPT.