Bug in NAPI...

[Ray Osborn]

on 2000/02/11 9:16 AM, Mark Koennecke at Mark.Koennecke at psi.ch wrote:

> You are right, this is a bug. Should we go for an own NXint32 or the HDF
> int32 for the arguments in order to fix it? It ocurred to no one so far
> that there might be 16-bit ints as default still out there. What was
> that environment exactly? Win32?
> 
> Mark 
> 

Brian says that he was able to fix the problem by simply assigning local
int32 arrays to the input arguments.  This would seem to have two
advantages.  Firstly, we would not require the calling program to use
non-standard variable types (except NXhandle etc), and secondly, it would
not require any rewriting of user code.  We can't use the name NXint32 in a
typedef since it's already in use as a constant.

Ray
-- 
Dr Ray Osborn                Tel: +1 (630) 252-9011
Materials Science Division   Fax: +1 (630) 252-7777
Argonne National Laboratory  E-mail: ROsborn at anl.gov
Argonne, IL 60439-4845