[MINC-development] MINC2 api - hidden treasures

[Andrew Janke]

Ack!  I wonder if our scheme of simply re-implementing MINC in HDF5
from scratch we mused about while I was in Montreal has any legs.. :)

a



On 29 November 2012 05:42, Vladimir S. FONOV <vladimir.fonov at gmail.com> wrote:
> Hello Everybody,
>
> I've been trying to make MINC2 API, NETCDF independent recently. And I have
> to say that it's been very frustrating experience so far.
>
> Occasionally I come across pieces of code like this:
> https://github.com/BIC-MNI/libminc/blob/master/libsrc2/volume.c#L98 (written
> in 2004):
>
>     if (volume->has_slice_scaling) {
>       /* TODO: Find the slowest-varying spatial dimension; that forms
>       * the basis for the image-min and image-max variables.  Right
>       * now this is an oversimplification!
>       */
>       ndims = volume->number_of_dims - 2;
>       dataspace_id = H5Screate_simple(ndims, hdf_size, NULL);
>     } else {
>       ndims = 0;
>       dataspace_id = H5Screate(H5S_SCALAR);
>     }
>
> Which means that MINC2 files created with MINC2 API are always assumed to
> have more then 2 dimensions if slice normalization is enabled. And only
> provide 2D slice normalization.
>
> --
> Best regards,
>
>  Vladimir S. FONOV ~ vladimir.fonov <at> gmail.com
> _______________________________________________
> MINC-development mailing list
> MINC-development at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-development