Dear Atsushi Togo,

Currently only VASP is supported out of the box.
A support of other main DFT codes would be great (in particular Quantum Espresso) for force calculations but this needs to be included at various points, e.g. in parsing the forces (QE saves them in the output trajectory) and in setting up the calculation.

Best regards,
Benedikt

An updated documentation is needed. The new docs will follow this general guidelines.

aiida-core should be specified as a dependency in setup.json

depending on the aiida version, consider something like

aiida-core>=0.12.3,<1.0.0

or

aiida-core>=1.0.0b6,<2.0.0

After using the phonopy plugin ,
I am using as input

BAND=0.0 0.0 0.0 0.3333 0.3333 0.0 0.5 0.5 0.0 0.0 0.0 0.0
BAND_LABELS=$\Gamma$ X M $\Gamma$
BAND_POINTS=100
FORCE_CONSTANTS=READ
NAC=.True.
HDF5=.True.

However, the output from PhonopyCalculation doesnt seem to parse the data correctly,
For example the labels arent parse
I inspected the band.hdf5 file and they are defined

with h5py.File('band.hdf5', 'r') as file:
... label = file['label']
... data = label[:]
... print(data)
...
[[b'$\Gamma$' b'X']
[b'X' b'M']
[b'M' b'$\Gamma$']]

The same information of the paths, arent parsed as well, i have two instead of three, and after inspecting the file manually indeed it has three.

please add the Framework :: AiiDA classifier and update the python language classifier (currently suggests that this plugin supports only python 3.5).

I believe this was meant to say 'install_requires':

Also, it would make sense to add the appropriate version of aiida-core to the requirements.

Related issue to #15, I plan to implement expose_inputs for each calculator (VASP for example) to pass input parameters.

I notice that the phono3py entry point breaks the naming convention that entry points provided by a plugin should start with the name of the plugin.
https://travis-ci.org/aiidateam/aiida-registry/jobs/591463873#L356

This may be a bit of a special case but it may still be worth renaming it to follow the standard...