Re: [Pyobjc-dev] NSDecimalNumber to the next level

[Ronald O. <ous...@ci...>]

On 9-mrt-04, at 19:29, Pierce T.Wetter III wrote:

>>>  I want to take NSDecimalNumber to the next level.
>>>
>>>  History:
>>>
>>>   1.0.
>>>
>>>    NSDecimalNumber is wrapped as identical to CFNumber, which turns  
>>> out not work.
>>>
>>>   1.1
>>>
>>>    NSDecimalNumbers are passed through untransformed, you can use:
>>>
>>>   dv=obj.doubleValue() to convert them to Python doubles, then:
>>>
>>>    NSDecimalNumber.numberWithDouble(dv) to convert them back. This  
>>> had to happen because the type information was getting lost, and the  
>>> values passed back to Objective-C were plain CFNumber classes, which  
>>> didn't implement NSDecimalNumber.decimalNumberByAdding_(dv), etc.
>>
>> The type info should not get lost, NSDecimalNumber and NSDecimal (the  
>> C type backing the NSDecimalNumber class) should work correctly as  
>> numbers, including save coercions. I explicitly disabled implicit  
>> coercion from NSDecimal{Number,} to and from float because you might  
>> loose precision that way.
>>
>> If this doesn't work you found a bug that should be fixed.
>
> Call me crazy, but I'd like to get this code to work:
>
> import objc
> import Foundation
> print  
> Foundation.NSDecimalNumber.zero().decimalNumberByAdding_(Foundation.NSD 
> ecimalNumber.zero())
> print Foundation.NSDecimalNumber.zero().decimalNumberByAdding_(1.0)
> That's what I would like to be able to do.

 >>> import Foundation
 >>>  
Foundation.NSDecimalNumber.zero().decimalNumberByAdding_(Foundation.NSDe 
cimalNumber.zero())
0
 >>> type(_)
<objective-c class NSDecimalNumber at 0xa0a061f4>
 >>> Foundation.NSDecimalNumber.zero().decimalNumberByAdding_(1.0)
0

The last one is a bug, it should raise TypeError just like:
 >>> Foundation.NSDecimal(0) + 1.0
Traceback (most recent call last):
   File "<stdin>", line 1, in ?
TypeError: unsupported operand type(s) for +: 'Foundation.NSDecimal'  
and 'float'

 From what I've seen from NSDecimal is simular to the type described in  
PEP 327 and the last time I saw discussion about this they seemed to  
favor explicit conversion from float for various reasons.

Ronald
--
X|support bv            http://www.xsupport.nl/
T:  +31 610271479       F:  +31 204416173