Dominik Gresch
Dominik Gresch
I think it might be useful to have Python bindings for the library mode (using something like f2py). What's your take on that?
Ok, I'll try to investigate why the ``.werr`` file doesn't appear in the sample case I had. Can you please assign the issue to me?
The problem is that ``io_error`` writes to the ``stdout`` unit in the non-MPI code [(see here)](https://github.com/wannier-developers/wannier90/blob/develop/src/io.F90#L244). In ``wannier_prog.F90``, ``stdout`` is set to the ``.werr`` file during ``param_read``, but later set...
> (b) try to automatically send it to a server somewhere. Maybe this should be a separate command (e.g. `verdi database telemetry upload`), as people might want to inspect the...
I think the "logical" place to control this would be the `is_valid_cache` property, see https://aiida.readthedocs.io/projects/aiida-core/en/latest/internals/engine.html#controlling-caching The main question is: should *all* calculations which have a "disappeared" `RemoteData` be considered invalid...
Great points @ramirezfranciscof! (1) yes, absolutely we need a way to check for `RemoteData` integrity (2) there is a [context manager](https://aiida.readthedocs.io/projects/aiida-core/en/latest/howto/run_codes.html#instance-level) to temporarily switch caching off, but that is a...
> This can then be used to see if the hierarchy is intact and on top of that check that the hashes of the files are still the same. My...
> use the hash of the previous PwCalculation step In this case, aren't the two hashes of `PwCJ_1` and `PwCJ_1'` the same? The "safe" option discussed previously of ignoring any...
> we could even schedule a periodic garbage collection call Hmm, I think that's what Python does anyway. Of course we then don't have control over when it happens. I...
> which aren't garbage-collected automatically They should be.. Python memory management has two components: - Reference counting: as soon as the ref counter hits zero, the object is destroyed -...