Re: more aggregate tests

On 20/09/10 04:40, Gregory Williams wrote:
> On Sep 18, 2010, at 9:17 AM, Andy Seaborne wrote:
>
>> :mixed2 is the sum of "2.2"^^xsd:decimal and "2e-1"^^xsd:double.  So the decimal is converted to a double and the result added together but ARQ gets 2.4000000000000004e0 - there is one bit of low end noise introduced by the addition being done on doubles of different exponent (this is Java double+double doing that step).  xsd:float, not double, works.
>
>
> Will float work consistently? If so, I can update the test to use it. If not, any thoughts on how we might test the execution of sum with datatype promotion in a portable manner?

Float might work (if it does, it's by chance).

Better to choose a double that has an exact representation (0.2, a 
fifth, does not, it's bit pattern as a double is 0x3FC999999999999A).

	Andy

>
> .greg
>