[Insight-developers] Failing Test after check in

[Bradley Lowekamp]

Bill,
	I have never streamed it myself. I think I can safely assume that the  
tests don't check for streaming too. However, the documentation says  
it can stream. Also filters which implement GenerateData can stream,  
consider a composite filter (combining multiple other filter). As long  
as pieces in the composited pipeline stream and the requested regions  
are respected the resulting filter should stream. Also the class I e- 
mailed you yesterday, which I started to use for testing and  
debugging, is close to a bare bones example non-thread streamer.
	MultiResolutionPyramidImageFilter is essentially multiple composite  
filters. So it should be able to stream.... theoretically :)

	Giving a more detailed look at the code, I can see why the regions  
are problematic. Internally it is computing the OutputInformation for  
each resolution. It would be trick to match the internal calculations  
of the resizing filter, and also error prone too. So when the code is  
actually executed it may not match.

Brad


On Dec 18, 2008, at 3:56 PM, Bill Lorensen wrote:

> Brad,
>
> I don't think that MultiResolutionPyramidImageFilter can stream
> because it implements GenerateData and not ThreadedGenerateData.
>
> Bill
>
> On Thu, Dec 18, 2008 at 3:09 PM, Bradley Lowekamp
> <blowekamp at mail.nih.gov> wrote:
>> Bill,
>> Thanks for fixing my bug :)
>> Unfortunately I have streaming on my mind. And I think you may  
>> created the
>> same bug, but in a different class. Wouldn't adding that line make  
>> is so
>> that the class does not stream any more?
>>
>> Brad
>> On Dec 18, 2008, at 1:22 PM, Bill Lorensen wrote:
>>
>> Hans,
>>
>> I fixed the problem. It was a tricky one. I recall that if an ITK
>> filter is reused and produces a different sized output, its request
>> region is not correct. I fixed the pyramid code to
>> UpdateLargestPossibleRegion() on the resample filter. Also, the
>> resample filter had a bug if the OutputStartIndex was not 0. The
>> itkMultiResolutionPDEDeformableRegistrationTest created an image with
>> a non-zero index. That caused the problem to surface in the resample
>> filter.
>>
>> Bill
>>
>> On Wed, Dec 17, 2008 at 2:00 PM, Bill Lorensen <bill.lorensen at gmail.com 
>> >
>> wrote:
>>
>> The other failures were different I think. I made this test into a
>>
>> regression test. The older failures were regression failures I
>>
>> believe.
>>
>> BTW, I'm almost ready to checkin the ShrinkFilter fixes.
>>
>> Bill
>>
>> On Wed, Dec 17, 2008 at 1:52 PM, Hans Johnson <hans- 
>> johnson at uiowa.edu>
>> wrote:
>>
>> Bill,
>>
>> This has been failing since at least December 15th nightly builds.  I
>>
>> committed my code last night at 9:30pm.
>>
>> I don't think that I am the originating culprit in this failure.   
>> With that
>>
>> said, I may be compounding the problem with my bug fixes as this  
>> filter
>>
>> almost certainly uses the itkMultiResolutionPyramidImageFilter.
>>
>>
>> Hans
>>
>>
>>
>>
>> On 12/17/08 11:47 AM, "Bill Lorensen" <bill.lorensen at gmail.com>  
>> wrote:
>>
>> Hans,
>>
>> The itkMultiResolutionPDEDeformableRegistrationTest is failing on  
>> some
>>
>> platforms after your recent checkins.
>>
>> http://www.cdash.org/CDash/testSummary.php?project=2&name=itkMultiResolutionPD
>>
>> EDeformableRegistrationTest&date=2008-12-17
>>
>> I think they are all failing in a similar way. If I recall correctly,
>>
>> this test uses images that have a non-zero starting index. This may  
>> be
>>
>> causing the problem.
>>
>> Bill
>>
>>
>>
>> Notice: This UI Health Care e-mail (including attachments) is  
>> covered by the
>> Electronic Communications Privacy Act, 18 U.S.C. 2510-2521, is  
>> confidential
>> and may be legally privileged.  If you are not the intended  
>> recipient, you
>> are hereby notified that any retention, dissemination,  
>> distribution, or
>> copying of this communication is strictly prohibited.  Please reply  
>> to the
>> sender that you have received the message in error, then delete  
>> it.  Thank
>> you.
>>
>>
>>
>>
>> _______________________________________________
>> Insight-developers mailing list
>> Insight-developers at itk.org
>> http://www.itk.org/mailman/listinfo/insight-developers
>>
>> ========================================================
>>
>> Bradley Lowekamp
>>
>> Lockheed Martin Contractor for
>>
>> Office of High Performance Computing and Communications
>>
>> National Library of Medicine
>>
>> blowekamp at mail.nih.gov
>>
>>

========================================================
Bradley Lowekamp
Lockheed Martin Contractor for
Office of High Performance Computing and Communications
National Library of Medicine
blowekamp at mail.nih.gov


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/mailman/private/insight-developers/attachments/20081218/2fc520fd/attachment-0001.htm>