search

materialsproject / emmet

Be a master builder of databases of material properties. Avoid the Kragle.
https://materialsproject.github.io/emmet/
Other
52 stars 64 forks source link

Bug when parsing LDA pseudopotential #847

Open Andrew-S-Rosen opened 11 months ago

Andrew-S-Rosen commented 11 months ago

Problem

Trying to read in a VASP run with an LDA (i.e. PAW but not PAW_PBE) pseudopotential causes a POTCAR parsing issue on emmet-core 0.69.6 and pmg 2023.10.4.

from emmet.core.tasks import TaskDoc

TaskDoc.from_directory('.')
File /global/common/software/matgen/rosen/miniconda/envs/quacc/lib/python3.10/sn, additional_fields, volume_change_warning_tol, **vasp_calculation_kwargs)
    488 if vasp_objects:
    489     included_objects = list(vasp_objects.keys())
    491 doc = cls.from_structure(
    492     structure=calcs_reversed[0].output.structure,
    493     meta_structure=calcs_reversed[0].output.structure,
    494     include_structure=True,
    495     dir_name=dir_name,
    496     calcs_reversed=calcs_reversed,
    497     analysis=analysis,
    498     transformations=transformations,
    499     custodian=custodian,
    500     orig_inputs=orig_inputs,
    501     additional_json=additional_json,
    502     icsd_id=icsd_id,
    503     tags=tags,
    504     author=author,
    505     completed_at=calcs_reversed[0].completed_at,
--> 506     input=InputDoc.from_vasp_calc_doc(calcs_reversed[-1]),
    507     output=OutputDoc.from_vasp_calc_doc(
    508         calcs_reversed[0],
    509         vasp_objects.get(VaspObject.TRAJECTORY),  # type: ignore
    510     ),
    511     state=_get_state(calcs_reversed, analysis),
    512     run_stats=_get_run_stats(calcs_reversed),
    513     vasp_objects=vasp_objects,
    514     included_objects=included_objects,
    515     task_type=calcs_reversed[0].task_type,
    516 )
    517 doc = doc.model_copy(update=additional_fields)
    518 return doc

File /global/common/software/matgen/rosen/miniconda/envs/quacc/lib/python3.10/s
    195 if xc:
    196     xc = xc.upper()
--> 198 pot_type, func = calc_doc.input.potcar_type[0].split("_")
    199 func = "lda" if len(pot_type) == 1 else "_".join(func)
    200 pps = Potcar(pot_type=pot_type, functional=func, symbols=calc_doc.input

ValueError: not enough values to unpack (expected 2, got 1)

I can't share a minimal example here due to proprietary reasons, but I can send one over Slack.

Proposed Solution

If the string label can't be split, assume LDA? I think?

Alternatives

No response