initial investigation into harmonizing particles and atoms

pyproject.toml

@@ -28,6 +28,11 @@ maintainers = [
     { name = "Joseph F. Rudzinski", email = "[email protected]" }
 ]
 license = { file = "LICENSE" }
+# dependencies = [
+#     "nomad-lab>=1.3.0",
+#     "matid>=2.0.0.dev2",
+#     "nomad-simulations@file:///home/bmohr/software/nomad-simulations",
+# ]
 dependencies = [
     "nomad-lab>=1.3.0",
     "matid>=2.0.0.dev2",
@@ -40,12 +45,13 @@ dependencies = [
 
 [project.optional-dependencies]
 dev = [
-    "mypy==1.0.1",
-    "ruff",
-    "pytest",
-    "pytest-timeout",
-    "pytest-cov",
-    "structlog",
+    'mypy==1.0.1',
+    'pytest>= 5.3.0, <8',
+    'pytest-timeout>=1.4.2',
+    'pytest-cov>=2.7.1',
+    'ruff>=0.6',
+    'structlog>=1.0',
+    'typing-extensions>=4.12',
 ]
 
 [tool.uv]

src/nomad_simulations/schema_packages/__init__.py

@@ -31,8 +31,8 @@ class NOMADSimulationsEntryPoint(SchemaPackageEntryPoint):
         description='Limite of the number of atoms in the unit cell to be treated for the system type classification from MatID to work. This is done to avoid overhead of the package.',
     )
     equal_cell_positions_tolerance: float = Field(
-        1e-12,
-        description='Tolerance (in meters) for the cell positions to be considered equal.',
+        12,
+        description='Decimal order or tolerance (in meters) for comparing cell positions.',
     )
 
     def load(self):

src/nomad_simulations/schema_packages/atoms_state.py

@@ -552,7 +552,16 @@ def normalize(self, archive: 'EntryArchive', logger: 'BoundLogger') -> None:
                 )
 
 
-class AtomsState(Entity):
+class State(Entity):
+    """
+    A base section to define the state information of the system.
+    """
+
+    def __init__(self, m_def: 'Section' = None, m_context: 'Context' = None, **kwargs):
+        super().__init__(m_def, m_context, **kwargs)
+
+
+class AtomsState(State):
     """
     A base section to define each atom state information.
     """

src/nomad_simulations/schema_packages/general.py

@@ -1,10 +1,17 @@
-from typing import TYPE_CHECKING
+#! TODO: Why is TYPE_CHECKING False?
+from typing import TYPE_CHECKING, List, Iterable, Union
 
-if TYPE_CHECKING:
-    from collections.abc import Callable
-
-    from nomad.datamodel.datamodel import EntryArchive
-    from structlog.stdlib import BoundLogger
+if not TYPE_CHECKING:
+    from nomad.datamodel.datamodel import (
+        EntryArchive,
+    )
+    from nomad.metainfo import (
+        Context,
+        Section,
+    )
+    from structlog.stdlib import (
+        BoundLogger,
+    )
 
 import numpy as np
 from nomad.config import config
@@ -217,7 +224,10 @@ def _set_system_branch_depth(
                 system_parent=system_child, branch_depth=branch_depth + 1
             )
 
-    def resolve_composition_formula(self, system_parent: ModelSystem) -> None:
+    #! Generalize from checks for atomic systems, error with CG input
+    def resolve_composition_formula(
+        self, system_parent: ModelSystem, logger: 'BoundLogger'
+    ) -> None:
         """Determine and set the composition formula for `system_parent` and all of its
         descendants.
 
@@ -226,7 +236,7 @@ def resolve_composition_formula(self, system_parent: ModelSystem) -> None:
         """
 
         def set_composition_formula(
-            system: ModelSystem, subsystems: list[ModelSystem], atom_labels: list[str]
+            system: ModelSystem, subsystems: list[ModelSystem], labels: list[str]
         ) -> None:
             """Determine the composition formula for `system` based on its `subsystems`.
             If `system` has no children, the atom_labels are used to determine the formula.
@@ -238,13 +248,15 @@ def set_composition_formula(
                 to the atom indices stored in system.
             """
             if not subsystems:
-                atom_indices = (
-                    system.atom_indices if system.atom_indices is not None else []
+                particle_indices = (
+                    system.particle_indices
+                    if system.particle_indices is not None
+                    else []
                 )
                 subsystem_labels = (
-                    [np.array(atom_labels)[atom_indices]]
-                    if atom_labels
-                    else ['Unknown' for atom in range(len(atom_indices))]
+                    [np.array(labels)[particle_indices]]
+                    if labels
+                    else ['Unknown' for atom in range(len(particle_indices))]
                 )
             else:
                 subsystem_labels = [
@@ -258,7 +270,7 @@ def set_composition_formula(
                     children_names=subsystem_labels
                 )
 
-        def get_composition_recurs(system: ModelSystem, atom_labels: list[str]) -> None:
+        def get_composition_recurs(system: ModelSystem, labels: list[str]) -> None:
             """Traverse the system hierarchy downward and set the branch composition for
             all (sub)systems at each level.
 
@@ -268,22 +280,17 @@ def get_composition_recurs(system: ModelSystem, atom_labels: list[str]) -> None:
                 to the atom indices stored in system.
             """
             subsystems = system.model_system
-            set_composition_formula(
-                system=system, subsystems=subsystems, atom_labels=atom_labels
-            )
+            set_composition_formula(system=system, subsystems=subsystems, labels=labels)
             if subsystems:
                 for subsystem in subsystems:
-                    get_composition_recurs(system=subsystem, atom_labels=atom_labels)
-
-        atoms_state = (
-            system_parent.cell[0].atoms_state if system_parent.cell is not None else []
-        )
-        atom_labels = (
-            [atom.chemical_symbol for atom in atoms_state]
-            if atoms_state is not None
-            else []
-        )
-        get_composition_recurs(system=system_parent, atom_labels=atom_labels)
+                    get_composition_recurs(system=subsystem, labels=labels)
+
+        # ! CG: system_parent.cell[0].particles_state instead of atoms_state!
+        labels = []
+        if system_parent.cell is not None:
+            labels = system_parent.cell[0].get('labels', logger=logger)
+
+        get_composition_recurs(system=system_parent, labels=labels)
 
     def normalize(self, archive: 'EntryArchive', logger: 'BoundLogger') -> None:
         super(Schema, self).normalize(archive, logger)
@@ -308,7 +315,7 @@ def normalize(self, archive: 'EntryArchive', logger: 'BoundLogger') -> None:
 
             if is_not_representative(model_system=system_parent, logger=logger):
                 continue
-            self.resolve_composition_formula(system_parent=system_parent)
+            self.resolve_composition_formula(system_parent=system_parent, logger=logger)
 
 
 m_package.__init_metainfo__()

src/nomad_simulations/schema_packages/model_method.py

@@ -544,7 +544,7 @@ def resolve_orbital_references(
             # If the child is not an "active_atom", the normalization will not run
             if active_atom.type != 'active_atom':
                 continue
-            indices = active_atom.atom_indices
+            indices = active_atom.particle_indices
             for index in indices:
                 try:
                     active_atoms_state = atoms_state[index]