[Insight-developers] ImageIO design and inner workings questi ons

[Kent Williams]

Today I determined that the problem I was seeing was this:  The original=20
author of the itk::AnalyzeImageIO class put a test into the CanRead and=20
CanWrite methods, that checks AFTER m_FileName is set, whether it is an
empty string.

The itk::ImageFileWriter template class does, in fact, set the image file=
 name=20
where it matters, right before Write gets called.

I'm intrigued, after further code spelunking, about the real purpose of t=
he
itk::ImageIOBase::WriteImageInformation() method.  This doesn't appear to=
 be=20
called from anywhere in the ITK source tree; itk::AnalyzeImageIO calls it=
 in=20
order to write out the Analyze .hdr file.  All the other classes in Code/=
IO=20
implement it with an empty function.

On Friday 28 March 2003 01:38 pm, Miller, James V (Research) wrote:
> Looking at the ImageFileWriter, it looks like the GenerateData()
> routine sets the filename on the ImageIO object right before the
> the it tells the ImageIO object to write the data.
>
>
> So I do not understand why the filename was not getting set on the
> Analyze ImageIO.
>