Satoshi Nakagawa
snaka****@infot*****
Sun Sep 9 09:06:53 JST 2007
I agreed.
I think it's good.
--
Satoshi Nakagawa
On 2007/09/09, at 2:39, Laurent Sansonetti wrote:
> I vote for this :)
>
> What do others think?
>
> Laurent
>
> On Sep 8, 2007, at 7:03 PM, Eloy Duran wrote:
>
>> Ok, so do we then settle on the following?
>>
>> #<NSCFString "foo">
>>
>> #<NSCFArray [#<NSCFString "foo">, #<NSCFNumber 99.99>]>
>>
>> #<NSCFNumber 99.99>
>> #<NSCFNumber true>
>> #<NSCFNumber 42>
>>
>> #<NSDictionary {#<NSCFString "foo"> => #<NSCFNumber 42>}>
>>
>> On 9/8/07, Laurent Sansonetti <lsans****@apple*****> wrote:
>>> I think we should keep the full class names. It's important for
>>> consistency, since all other #inspect methods return them. We could
>>> nevertheless drop the OSX prefix in all #inspect methods.
>>>
>>> The IDs are interesting to determine if 2 objects of the same value
>>> are different instances, or the same object. Originally I was
>>> thinking
>>> about keeping them too, but I revised by judgment. ObjC doesn't
>>> reveal
>>> them in the primitive types [-description] methods, so we shouldn't
>>> need too. One can still call #ocid to determine the ObjC address.
>>>
>>> Laurent
>>>
>>> On Sep 8, 2007, at 1:12 PM, Eloy Duran wrote:
>>>
>>>> Hey Satoshi-san,
>>>>
>>>> I think we should at least show the full classname, so
>>>> OSX::NSCFString etc.
>>>>
>>>> The id's are indeed a point of debate, because also with ruby's
>>>> String, Array, Hash, and Numeric the #inspect method doesn't show
>>>> the
>>>> id....
>>>>
>>>> Eloy
>>>>
>>>> On 9/8/07, Satoshi Nakagawa <snaka****@infot*****> wrote:
>>>>> Hi.
>>>>>
>>>>> I prefer shorter form.
>>>>>
>>>>> <NSString "foo">
>>>>> <NSNumber 99.99>
>>>>> <NSArray [<NSString "foo">, <NSNumber 99.99>]>
>>>>> <NSDictionary {<NSString "foo"> => <NSNumber 42>}>
>>>>>
>>>>> or
>>>>>
>>>>> <NS "foo">
>>>>> <NS 99.99>
>>>>> <NS [<NS "foo">, <NS 99.99>]>
>>>>> <NS {<NS "foo"> => <NS 42>}>
>>>>>
>>>>> Because these types are value types, so we don't need to
>>>>> know their ids.
>>>>>
>>>>> --
>>>>> Satoshi Nakagawa