[netCDF #SRY-526862]: hdf / netcdf

Oops, on re-reading the HDF5 known-problems web page, I think I 
need to correct what I wrote:
> ...  writing a file with
> any object that has more than 8 attributes on a little-endian machine
> and then trying to read it on a big endian machine may cause a
> segmentation violation, for example. ...

got the big- and little-endian reversed.  I thought my tests showed 
there was a problem either way, but the known-problems page

  http://www.hdfgroup.org/HDF5/release/known_problems/

claims it's only when writing on a big-endian machine that the file is
bad on little-endian machines.  I'm going to check this, and will let you 
know if it's wrong.

--Russ

Russ Rew                                         UCAR Unidata Program
address@hidden                      http://www.unidata.ucar.edu



Ticket Details
===================
Ticket ID: SRY-526862
Department: Support netCDF
Priority: Normal
Status: Closed