[Nexus] Switching to a debate about NeXus + Python

[Darren Dale]

On Mar 10, 2010, at 2:48 AM, Wellenreuther, Gerd wrote:

> Hi Ray,
> 
> I would like to talk about that module Paul Kienzle produced in more detail. The reason for that: A lot of scientists consider Python as a very good environment for scripting/programming purposes, and having a top-level python module supporting NeXus would be a very good thing.
> 
> Recently, I have tried phynx written by Darren Dale, which is already providing support for NeXus. I know of h5py and pytables, which are capable of writing HDF5-files directly, but I would like to use a module which does most of the tiring and repetitive setting of the NeXus-internals like NXclass, signals, axes etc. itself.

Phynx is currently a step in that direction, but it may be the case that Ray's project is further along since it is built upon the official nexus API.

> I personally would also like to split the task of defining where which data during acquisition time has to go in the NeXus-file, from the task of actually writing the file. E.g. I built another module on top of phynx (which is itself on top of h5py, Darren, please correct me).

Right, phynx is built on h5py because it has an intuitive interface, is thread-safe, and provides a near-complete wrapping of the hdf5 library.

Darren

---
Darren S. Dale, Ph.D.
Staff Scientist
Cornell High Energy Synchrotron Source
Cornell University
275 Wilson Lab
Rt. 366 & Pine Tree Road
Ithaca, NY 14853

darren.dale at cornell.edu
office: (607) 255-3819
fax: (607) 255-9001
http://www.chess.cornell.edu