On Tue, May 4, 2010 at 5:05 AM, Andy Buckle <[email protected]> wrote:
> follows from: 
> http://octave.1599824.n4.nabble.com/dicom-support-tt2122699.html#a2122699
>
> Following from my comment on storing the dicom dictionary. I tried a
> few options;

One of the issues is that if we want full M compatibility with
dicomdict() we should support modifying the dictionary in run-time
including changing value representations; however, it is questionable
whether M's approach of allowing all modifications even to the portion
of the dictionaries specified by the standard is a good idea. I
hesitate to pronounce this as obvious law to guide implementation
because I know from experience that vendors can both ship with bugs
and get ahead of the standards.  I think we need to determine how much
of the gdcm dictionaries can be modified in run-time versus creating a
new Part3.xml or recompiling gdcm source. Tag mapping isn't too
difficult, it's what is involved overriding of VR's and adding new
private tags or replacing existing private tags that I don't know
about yet. I do like that gdcm already includes and extensive
repository of known private tags.


--judd

------------------------------------------------------------------------------
_______________________________________________
Octave-dev mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/octave-dev

Reply via email to