[Insight-developers] itkTimeStamp Test Failures Mistery

[Tom Vercauteren]

Hi Luis,

I should be able to modify the test on Monday or Tuesday. Is this OK for you?

Tom

On Fri, Feb 20, 2009 at 18:50, Luis Ibanez <luis.ibanez at kitware.com> wrote:
> Tom,
>
> Let's do that then.
>
> Would you like to modify the test ?
>
> It will be nice to put this one down,
> since we have 8 days left for the
> release deadline.
>
>
> Please let us know.
>
>
>    Thanks
>
>
>       Luis
>
>
> -------------------------
> Tom Vercauteren wrote:
>>
>> Hi Luis,
>>
>> I also feel that adding some documentation in the TimeStamp about this
>> behavior and modifying the TimeStampTest should be enough.
>>
>> Tom
>>
>> On Fri, Feb 20, 2009 at 18:21, Luis Ibanez <luis.ibanez at kitware.com>
>> wrote:
>>
>>> Brad
>>>
>>> That modification sounds reasonable to me.
>>>
>>> Unless somebody can come up with a typical use that actually
>>> require TimeStamp::Modified() to be thread-safe...
>>>
>>>
>>>  Luis
>>>
>>>
>>> --------------------------
>>> Bradley Lowekamp wrote:
>>>
>>>> So should we change the test so that each thread has it own itkTimeStamp
>>>> class. There by just testing that the class is reentrant. And then
>>>> specify
>>>> in the TimeStamp documentation class that it is only re-entrant.
>>>> I think that the test should then pass.  And it'll test the needed
>>>> functionality. Anyone object?
>>>>
>>>> Brad
>>>>
>>>
>>
>