Re: [h5md-user] fix remaining imprecisions - particle position

[Felix Höfling]

Am 10.12.2013, 11:54 Uhr, schrieb Felix Höfling <address@hidden>:

> * For "position", I can't recall the criterion whether the data are  
> interpreted as "wrapped" or "unwrapped". This should be stated clearly.  
> (BTW, would "folded coordinates" be better instead of "wrapped  
> coordinates"?)

Happy New Year to everybody following the list!

The interpretation of the position data needs clarification still. I have  
made a suggestion in commit 0989a65, which reads now:

`position`
:   An element that describes the particle positions as coordinate vectors  
of
    `Float` or `Integer` type. If the element `image` is present as well,  
the
    data are interpreted as periodically reduced coordinates along the  
periodic
    directions of the simulation box. Otherwise, the data indicate the  
absolute
    position in space.

`image`
:   An element that represents the periodic image of the box in which each
    particle is actually located and allows one to unfold periodically  
reduced
    positions. For the case of time-dependent data, the `image/value`  
dataset
    is of the same shape as `position/value` and is either of `Float` or
    `Integer` type.

    For instance, given a cuboid box with periodic boundaries, let $\vec  
r_i$
    be the reduced position of particle $i$ taken from `position`, $\vec  
a_i$
    its image vector from `image`, and $\vec L$ the space diagonal of the  
box,
    then component $j$ of the absolute position $\vec R_i$ is given by
    $R_{ij} = r_{ij} + L_j a_{ij}$.

Feedback is most welcome!

Best regards,

Felix