On Wed, May 26, 2010 at 12:41 PM, John LaBanca <[email protected]> wrote:

>  Do you agree that we would still have a NumberParser even with the
>> renderers?
>>
>
> Uh, what?
>
>
> *NumberParser => NumberCell
>

I'm not sure I do, actually. It may be easy enough to do new
ValueCell<Number>(new NumberParser())  that we could drop it. But the point
is moot until gwt.input hits trunk, and ValueCell exists.

>
> Thanks,
> John LaBanca
> [email protected]
>
>
> On Wed, May 26, 2010 at 3:39 PM, Ray Ryan <[email protected]> wrote:
>
>> On Wed, May 26, 2010 at 12:29 PM, John LaBanca <[email protected]>wrote:
>>
>>> Hmmm... are you suggesting a ValueCell/ValueInputCell that defers to a
>>> Renderer/Parser?  Assuming you mean that we would have subclasses for each
>>> type as we do with ValueBox, I agree.  But that means that we'll have a
>>> DateCell and NumberCell anyway, so this patch can go in and we can add a
>>> ValueCell super class later.
>>>
>>
>> Yes. So I guess this is an interesting conversation but need not hold up
>> this CL.
>>
>>>
>>> That being said, I'm wary of having non-internationalized parsers such as
>>> DoubleParser.  If I'm parsing user input, I probably want to use
>>> NumberFormat.parser, which will handle doubles in the form 1,22 (unless
>>> Double.parseDouble does that).
>>>
>>
>> No argument. DoubleParser was a thoughtless last second hack, and is
>> probably wrong.
>>
>>>
>>> Do you agree that we would still have a NumberParser even with the
>>> renderers?
>>>
>>
>> Uh, what?
>>
>>>
>>> Thanks,
>>> John LaBanca
>>> [email protected]
>>>
>>>
>>>
>>> On Wed, May 26, 2010 at 3:04 PM, <[email protected]> wrote:
>>>
>>>> LGTM
>>>>
>>>>
>>>> http://gwt-code-reviews.appspot.com/568801/show
>>>>
>>>
>>>  --
>>> http://groups.google.com/group/Google-Web-Toolkit-Contributors
>>>
>>
>>
>

-- 
http://groups.google.com/group/Google-Web-Toolkit-Contributors

Reply via email to