Re: [h5md-user] Unit attribute versus non-dimensionless quantities

[Pierre de Buyl]

Dear all,

For those interested in the compact storage layout for datasets, I added a
routine "create_compact_dataset" to pyh5md.utils. It was not possible, from what
I understand, to use the routines provided by h5py for dataset creation with a
compact storage layout.

Currently, pyh5md can store fixed-in-time boxes. This is currently the only use
case for compact storage layout in pyh5md. A box is used in example
"random_walk_1d.py". 

Cheers,

Pierre

On Mon, Sep 09, 2013 at 08:36:57AM -0400, Peter Colberg wrote:
> Hi Konrad,
> 
> On Mon, Sep 09, 2013 at 12:37:48PM +0200, Konrad Hinsen wrote:
> > BTW, I wonder if storing data in attributes is really worth the
> > trouble it generates, considering units or in fact any other kind of
> > annotation.  If size is the motivation, then compact datasets can
> > provide nearly the same advantages without the inconvenience.
> 
> Thank you, that is a very helpful hint!
> 
> Compact datasets [1] are applicable to data of 30 kB or less, which is
> stored in the header block of the object. Compact datasets are created
> by setting the appropriate layout on the dataset creation property
> list using H5Pset_layout [2].
> 
> [1] http://www.hdfgroup.org/HDF5/doc/UG/UG_frame10Datasets.html
> [2] http://www.hdfgroup.org/HDF5/doc/RM/RM_H5P.html#Property-SetLayout
> 
> This allows us to mandate the storage of time-independent data using
> datasets, e.g., for fixed-size box information. In consequence, a
> “unit” attribute would be attachable to all data in the specification.
> 
> Peter
>