Hello,
I am populating a financial application, that requires a high degree of accuracy for obvious reasons...
I have a required field of Float datatype, that has some data quality issues at the source, in that the source value is coming as a zero...however, zero is not an acceptable value for the application.
I have tried replacing zero with "0.00000001" or 0.99E-99, but DataStage seems to reset the value to 0 again. I have tested my (simple) expression with other values (such as "9999999999" for a test), and it works correctly...
NOTE ...I misquoted the number I used...it was "0.00000000000000000001" and this reverts to 0
Any ideas? What IS the minimum value that can be populated to a FLOAT datatype without being defaulted back to zero? This seems like strange behaviour....
Thanks
W
FLOAT Datatype; Minimum Value?
Moderators: chulett, rschirm, roy
-
- Premium Member
- Posts: 40
- Joined: Thu Mar 30, 2006 6:30 am
- Location: Near Geneva, Switzerland
- Contact:
FLOAT Datatype; Minimum Value?
Last edited by wwalker on Thu Jul 27, 2006 7:14 am, edited 2 times in total.
Wade Walker
Which is your database?
I tried with Sql Server with the value
I got
The datatype of the column is float.
I tried with Sql Server with the value
Code: Select all
.0000000000000000001
in the table.1.0000000000000001E-19
The datatype of the column is float.
Success consists of getting up just one more time than you fall.
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
Maximum precision of numeric data within DataStage server jobs is, by default, 15 significant digits. This default can be overridden with the EXACTNUMERIC configuration parameter in uvconfig.
The PRECISION statement in BASIC affects the maximum number of digits that will appear to the right of the decimal place when a numeric item of data is prepared for presentation (converted to string).
The PRECISION statement in BASIC affects the maximum number of digits that will appear to the right of the decimal place when a numeric item of data is prepared for presentation (converted to string).
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
-
- Premium Member
- Posts: 224
- Joined: Tue Sep 24, 2002 7:32 am
- Location: Denver, CO USA
I believe you are hitting the setting called EXACTNUMERIC. It is in the uvconfig file. The default is 15 and I would not recommend changing it. The comment in the uvconfig file says "Specifies the number of digits of precision before rounding occurs".
Try using a value that is shorter than your exact_numeric setting, like 0.0001.
John
Try using a value that is shorter than your exact_numeric setting, like 0.0001.
John