[Insight-developers] itkDenseFrequencyContainer.txx

[Stephen R. Aylward]

Hi Jim,

Yes, it should be removed.

The Dense and Sparse frequency containers were modified in release 2.2. 
   We kept the class' API consistent, but the use of itk::Histogram (and 
related classes) now requires linking with the statistics library.  The 
template parameter in frequency containers weren't being applied 
consistently - resulting in errors if people changed it from the default 
value.   Since the other revisions to statistics library (to handle 
run-time measurement vector lengths) also required people to change 
their programs to use related classes in 2.2, we decided that it was 
appropriate timing to make this change...  Looking back, we should have 
had broader conversation on this.   I totally agree - we need a backward 
compatibility policy!

Sorry for wasting your time.   Yes - linking is part of the API....

Stephen


Miller, James V (Research) wrote:
> Should itkDenseFrequencyContainer.txx be removed from the repository?  It looks like DenseFrequencyContainer was changed from a templated class to a non-templated class and this file was orphaned.
> 
> On a side note. I just banged my head against a wall for a couple of hours trying to figure out why code that built against an ITK from early August would not build against ITK 2.2.  While the change to make DenseFrequencyContainer non-templated did not result in any compiler
> error in my existing code base (DenseFrequencyContainer had resonable defaults for its template parameters before), my program would no longer link.  Since DenseFrequencyContainer used to be templated, I did not have to link against ITKStatistics.  Now that is non-templated, my program needs to link against ITKStatistics.
> 
> Not a huge deal but it took me a while to deduce the link errors.
> 
> This is yet another type of backward compatibility issue that we need to monitor.
> 
> Jim Miller 
> _____________________________________
> Visualization & Computer Vision
> GE Research
> Bldg. KW, Room C223
> 1 Research Circle, Schenectady NY 12309-1027
> 
> millerjv at research.ge.com <mailto:millerjv at research.ge.com>
> (518) 387-4005, Dial Comm: 8*833-4005
> Cell: (518) 505-7065, Fax: (518) 387-6981
> 
> 
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Insight-developers mailing list
> Insight-developers at itk.org
> http://www.itk.org/mailman/listinfo/insight-developers

-- 
===========================================================
Dr. Stephen R. Aylward
Associate Professor of Radiology
Adjunct Associate Professor of Computer Science and Surgery
http://caddlab.rad.unc.edu
aylward at unc.edu
(919) 966-9695