Update H5MD parser and schema (#143)

* Update H5MD parser and schema

* Fix linting errors

* Fix linting and formatting (imports)

* ruff

* rebase correction

---------

Co-authored-by: jrudz <rudzinski@mpip-mainz.mpg.de>

Author: amirgolp

src/nomad_simulation_parsers/parsers/h5md/parser.py

@@ -8,13 +8,10 @@
 from nomad.parsing.parser import MatchingParser
 from nomad.units import ureg
 from nomad.utils import get_logger
-
-# from nomad_simulations.schema_packages.workflow import molecular_dynamics
 from structlog.stdlib import BoundLogger
 
 from nomad_simulation_parsers.parsers.utils.mdparserutils import MDParser
 from nomad_simulation_parsers.schema_packages import h5md
-from nomad_simulation_parsers.schema_packages.h5md import MolecularDynamics, Simulation
 
 LOGGER = get_logger(__name__)
 
@@ -504,10 +501,10 @@ def write_to_archive(self) -> None:
         # TODO consider using a single parser for the whole archive
         # create metainfo parsers
         self.simulation_parser.annotation_key = h5md.HDF5_KEY
-        simulation_data = Simulation()
+        simulation_data = h5md.Simulation()
         self.simulation_parser.data_object = simulation_data
         self.workflow_parser.annotation_key = h5md.HDF5_KEY
-        workflow_data = MolecularDynamics()
+        workflow_data = h5md.MolecularDynamics()
         self.workflow_parser.data_object = workflow_data
 
         # map from h5 source to metainfo target

src/nomad_simulation_parsers/parsers/lammps/file_parsers.py

@@ -435,7 +435,7 @@ def __init__(self) -> None:
 
     def init_quantities(self) -> None:
         def str_op(val: str) -> str | list[str]:
-            val = val.split('#',  1)[0]
+            val = val.split('#', 1)[0]
             val = re.sub(f'&{RE_N}+', ' ', val)
             val = val.split()
             val = val if len(val) > 1 else val[0]

src/nomad_simulation_parsers/parsers/lammps/parser.py

@@ -553,7 +553,7 @@ def _create_custom_parser(
         else:
             # TODO: Assumes the extension is always a valid lammps dump format, improve
             # Fallback to file extension
-            file_type = traj_file.split('.', 1)[-1]
+            file_type = traj_file.rsplit('.', 1)[-1]
 
         # TODO: add support for other LAMMPS dump file formats (https://docs.lammps.org/dump.html)
         if file_type == 'dcd' or file_type == 'xyz' and data_file:

src/nomad_simulation_parsers/schema_packages/h5md.py

@@ -266,6 +266,13 @@ class CustomProperty(physical_property.PhysicalProperty):
     )
     add_mapping_annotation(name, HDF5_KEY, '.name')
 
+    contribution_type = Quantity(
+        type=str,
+        description="""
+        Type of contribution to the physical property.
+        """,
+    )
+
     value = Quantity(
         type=np.dtype(np.float64),
         shape=[],

tests/parsers/test_h5md_parser.py

@@ -61,9 +61,9 @@ def assert_systems(archive: EntryArchive) -> None:
 
     assert sec_systems[2].positions[80][1].to('angstrom').magnitude == approx(28.748762)
     assert sec_systems[2].velocities[50][2].to('angstrom/ps').magnitude == approx(400.0)
-    assert sec_systems[3].lattice_vectors[2][2].to(
-        'angstrom'
-    ).magnitude == approx(68.22318)
+    assert sec_systems[3].lattice_vectors[2][2].to('angstrom').magnitude == approx(
+        68.22318
+    )
     assert sec_systems[3].periodic_boundary_conditions == [True, True, True]
     assert sec_systems[0].bond_list[200][0] == 198
     assert sec_systems[0].dimensionality == 3