Hi,
I am trying to pull the FLOAT column from my source (SQL SERVER) through DRS/ODBC connector. Value in source is '7.59999990463257'. But once I extract this value into dataset through datastage extraction job, I am able to see '7.6' if I use 'double' datatype. Busiess wants the complete value from source to target. My target is oracle and the datatype in oracle for this column is 'NUMBER'. I tried with different datatypes available in datastage, but not getting exact value like source. And one more thing, scale length is not fixed. As it is mentioned as FLOAT, 'dot' can come at any place. I am able to get exact value of source if I use 'decimal(38,14) in datastage. But we can not define the scale length as '14' as few source records are having 15 digits after 'dot'.
Could you please assist on this?
I tried with all possible datatypes in datastage like real, double, float, varchar but no use.
Scale value is missing for FLOAT column
Moderators: chulett, rschirm, roy
-
- Premium Member
- Posts: 425
- Joined: Sat Nov 19, 2005 9:26 am
- Location: New York City
- Contact:
Hi Pete,
On the source try defining the column as dfloat, and the oracle target as decimal with the proper precision and scale to fit your values. You would need to use dfloattodecimal function
While checking the values out of your source take in consideration that the values will be affected by the default set in the APT_DECIMAL_INTERM_PRECISION and APT_DECIMAL_INTERM_SCALE.....check what's the default of the scale one in the DS project and adjust accordingly
Regards
On the source try defining the column as dfloat, and the oracle target as decimal with the proper precision and scale to fit your values. You would need to use dfloattodecimal function
While checking the values out of your source take in consideration that the values will be affected by the default set in the APT_DECIMAL_INTERM_PRECISION and APT_DECIMAL_INTERM_SCALE.....check what's the default of the scale one in the DS project and adjust accordingly
Regards
Julio Rodriguez
ETL Developer by choice
"Sure we have lots of reasons for being rude - But no excuses
ETL Developer by choice
"Sure we have lots of reasons for being rude - But no excuses