Re: [h5md-user] H5MD 1.1 roadmap

[Pierre de Buyl]

On Sat, Jun 06, 2015 at 11:54:20PM +0200, Pierre de Buyl wrote:
> I attach a code that test all of what is proposed (except for charge, but 
> there
> is unknown on that side). There is also the resulting H5MD file.
> 
> The choices I made are:
> - proposal 100: use of mixed datasets and attributes
> - proposal 101: not updating for neighbor lists, too complex
> - proposal 102: as is
> - proposal 103: as is but the reference to 101 must be updated accordingly

I put my development/test code here: https://github.com/pdebuyl/h5md_dev This
test code was intented to check the application of H5MD 1.1 but I think that it
will also form the basis of the pyh5md code. The program "run.py" generates
"example_for_1.1.h5".

The only unknown that remains is: where to put "particle lists". They can go in
"connectivity" as far as I am concerned.



> 
> Pierre
> 
> PS: the deadline was june, was it?
>