DSXchange

Inputstring=5.9999999999999999999
expectoutoput(decimal,26,8)=5.99999999

but i am getting 6.00000000 (decimal,26,8)

how to acevied this ia m using stringtodecimal

Please help

Basically i don't want to round the input value.
Is there any environment variable to disable?
[quote][/quote]

There are four rtype arguments you can use with decimal processing; ceil, floor, round_inf and trunc_zero. The default is trunc_zero.

Read about them in Appendix B of the Parallel Job Developer's Guide for more information.

Hi,
I am not able to recollect the exact parameter name but i think it is:
APT_DECIMAL_INTER_ROUND_MODE... or
APT_DEC_INTERM_ROUND_MODE

You can check the variable :
APT_DECIMAL_INTER_ROUND_MODE=None.

I am not able to check the result since right now i do not have the acess to DS.
Hope this will help you out!

Hi there, Thanks for possible solutions,

Sol1:
I can applied APT_DECIMAL_INTERM_PRICISION TO NONE
-But the same results.

Sol2:
i know there are ceil,floor, round_inf and trunc_zero(default)
-But all there are meant to say how to round the input value.

But i don't want to round the input

input(varchar300)=5.99999999999
expected output(Decimal26,8)=5.99999999 but its rounding to 6.00000000

By now with several experiements i am using

AsDouble(Input) or DfloatToDecimal(AsDouble(pTest))
These would still round but to very less extend

Any other trick would help.

Did you try floor as the rtype?

Thanks Ray,

'floor' option has solved one requirement i.e for not to round the values

but this will not cater for exponential values.
inputvalves=5.99999999999 and 9.9999999999999995e-07

using StringToDecimal(inputvalues,'floor') does not work for expo values

as a test job i am handling expo values as AsDouble(inputvalues)

Please advice to handle for both types in inputs.

Thanks in advance

Keep experimenting. Curiously enough, we don't know all the answers. There are known unknowns, and there are unknown unknowns.

Post your results.

Well.
This works DfloatToDecimal(AsDouble(Input))
even for exponential values.

But still does some minor rounding

Thanks