Bug in NAPI...

[Ray Osborn]

on 2000/02/10 10:08 PM, Brian Tieman at tieman at aps.anl.gov wrote:

> I'm developing on Win95/98/NT and on all three platforms, ints are supposed to
> be 32 bit.  Also, I've never had a problem with this before.
> 
> So, I looked into the problem further and I think there may be a bug in my
> compiler (Inprise's CBuilder).  CBuilder will still build 16 bit code if
> requested (16 bit code will still run on 95/98 in most cases) although I've
> never intentionally tried it.  Doing some further tests today, I've been
> unable to recreate the problem under any circumstance.  My guess is something
> confused my compiler to start generating 16 bit code which caused the problem
> to surface.
> 

Nevertheless, people are usually free to choose what length a standard int
will be with a compiler switch, and may have legitimate reasons for choosing
16 bits.  We should still produce a version which works whatever the size.
We will have to check the code for all cases in which integers are typecast
when passed to the HDF library.

Ray

P.S. NeXus at anl.gov and NeXus at anpns1.pns.anl.gov are the same address, so
it's not necessary to Cc to both of them.

-- 
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