- Fix plugin setup

- Distinguish between parser reference and entry point reference - TODO: fix `m_annotations` overloading
FAIRmat-NFDI · Jul 3, 2024 · 5de6f3f · 5de6f3f
1 parent a605891
commit 5de6f3f
Show file tree

Hide file tree

Showing 3 changed files with 50 additions and 130 deletions.
diff --git a/pyproject.toml b/pyproject.toml
@@ -74,7 +74,7 @@ select = [
     "UP",
     # isort
     "I",
-    # pylint 
+    # pylint
     "PL",
 ]
 
@@ -108,8 +108,7 @@ package-dir = { "" = "src" }
 where = ["src"]
 
 [project.entry-points.'nomad.plugin']
-myparser = "nomad_parser_vasp.parsers:myparser"
-mypackage = "nomad_parser_vasp.schema_packages:mypackage"
+xml_parser = "nomad_parser_vasp.parsers:xml_entry_point"
 
 
 [tool.cruft]

diff --git a/src/nomad_parser_vasp/parsers/__init__.py b/src/nomad_parser_vasp/parsers/__init__.py
@@ -2,17 +2,17 @@
 from pydantic import Field
 
 
-class MyParserEntryPoint(ParserEntryPoint):
+class VasprunXMLEntryPoint(ParserEntryPoint):
     parameter: int = Field(0, description='Custom configuration parameter')
 
     def load(self):
-        from nomad_parser_vasp.parsers.myparser import MyParser
+        from nomad_parser_vasp.parsers.xml_parser import VasprunXMLParser
 
-        return MyParser(**self.dict())
+        return VasprunXMLParser(**self.dict())
 
 
-myparser = MyParserEntryPoint(
-    name='MyParser',
-    description='Parser defined using the new plugin mechanism.',
-    mainfile_name_re='.*\.myparser',
+xml_entry_point = VasprunXMLEntryPoint(
+    name='VasprunXML Parser',
+    description='Parser for VASP output in XML format.',
+    mainfile_name_re='.*vasprun\.xml.*',
 )
diff --git a/src/nomad_parser_vasp/parsers/xml_parser.py b/src/nomad_parser_vasp/parsers/xml_parser.py
@@ -1,123 +1,56 @@
-from typing import (
-    TYPE_CHECKING,
+from nomad.datamodel.datamodel import (
+    EntryArchive,
+)
+from structlog.stdlib import (
+    BoundLogger,
 )
-
-if TYPE_CHECKING:
-    from nomad.datamodel.datamodel import (
-        EntryArchive,
-    )
-    from structlog.stdlib import (
-        BoundLogger,
-    )
 
 from nomad.config import config
+from nomad.parsing import MatchingParser
 from nomad.parsing.file_parser.mapping_parser import (
     MappingAnnotationModel,
     MetainfoParser,
     XMLParser,
 )
-from nomad_simulations_plugin.numerical_settings import KMesh
+from nomad_simulations.schema_packages.general import Simulation, Program
+from nomad_simulations.schema_packages.model_system import (
+    ModelSystem,
+    AtomicCell,
+)
+from nomad_simulations.schema_packages.model_method import (
+    DFT,
+    XCFunctional,
+)
+from nomad_simulations.schema_packages.outputs import ElectronicEigenvalues
+from nomad_simulations.schema_packages.numerical_settings import KMesh
 
-configuration = config.get_plugin_entry_point('nomad_parser_vasp.parsers:myparser')
+configuration = config.get_plugin_entry_point(
+    'nomad_parser_vasp.parsers:xml_entry_point'
+)
 
 # TODO move to schema parser and prepare all relevant annotations for every file format
 
+
 # ! vasprun.xml has many meta fields, explaining field semantics
-Simulation.m_annotations = dict(xml=MappingAnnotationModel(path='modeling'))
-Simulation.program.name.m_annotations = dict(
-    xml=MappingAnnotationModel(path='modeling.generator.i[?"@name"="program"]')
-)
-Simulation.program.version.m_annotations = dict(
-    xml=MappingAnnotationModel(path='modeling.generator.i[?"@name"="version"]')
-)
-# ? compilation mode?
-Simulation.program.compilation_host.m_annotations = dict(
-    xml=MappingAnnotationModel(path='modeling.generator.i[?"@name"="platform"]')
-)
-KMesh.grid.m_annotations = dict(
-    xml=MappingAnnotationModel(
-        path='modeling.kpoints.generation.v[?"@name"="divisions"]'
-    )
-)
-KMesh.offset.m_annotations = dict(
-    xml=MappingAnnotationModel(path='modeling.kpoints.generation.v[?"@name"="shift"]')
-)
-KMesh.high_symmetry_points.m_annotations = dict(
-    xml=MappingAnnotationModel(path='modeling.kpoints.varray.v[?"@name"="kpointlist"]')
-)
-KMesh.weights.m_annotations = dict(
-    xml=MappingAnnotationModel(path='modeling.kpoints.varray.v[?"@name"="weights"]')
-)
-dft_path = 'modeling.separator[?"@name"="electronic exchange-correlation"]'
-# DFT.xc_functionals.m_annotations = dict(xml=MappingAnnotationModel(path=dft_path))
-DFT.exact_exchange_mixing_factor.m_annotations = dict(
-    xml=MappingAnnotationModel(
-        operator=(
-            lambda mix, cond: mix if cond else 0,
-            [dft_path + '.i[?"@name"="HFALPHA"]', dft_path + '.i[?"@name"="LHFCALC"]'],
-        )
-    )  # TODO convert vasp bool
-)
-XCFunctional.libxc_name.m_annotations = dict(
-    xml=MappingAnnotationModel(
-        operator=(convert_xc, [dft_path + '.i[?"@name"="GGA"]'])  # add LDA & mGGA
-    )
-)
-# ? target <structure name="initialpos" > and <structure name="finalpos" >
-AtomicCell.positions.m_annotations = dict(
-    xml=MappingAnnotationModel(
-        path='calculation.structure.varray[?"@name"="positions"]'
-    )
-)
-"""
-forces.m_annotations = dict(
-    xml=MappingAnnotationModel(path='calculation.structure.varray[?"@name"="forces"]')
-)
-stress.m_annotations = dict(
-    xml=MappingAnnotationModel(path='calculation.structure.varray[?"@name"="stress"]')
-)
-"""
-AtomicCell.lattice_vectors.m_annotations = dict(
-    xml=MappingAnnotationModel(
-        path='calculation.structure.crystal.varray[?"@name"="basis"]'
-    )
-)
-"""
-cell_volume.m_annotations = dict(
-    xml=MappingAnnotationModel(path='calculation.structure.crystal.i[?"@name"="volume"]')
-)
-reciprocal_lattice_vectors.m_annotations = dict(
-    xml=MappingAnnotationModel(path='calculation.structure.crystal.varray[?"@name"="rec_basis"]')
-)
-total_free_energy.m_annotations = dict(
-    xml=MappingAnnotationModel(path='calculation.energy.i[?"@name"="e_fr_energy"]')
-)
-total_internal_energy.m_annotations = dict(
-    xml=MappingAnnotationModel(path='calculation.energy.i[?"@name"="e_0_energy"]')
-)
-"""
-ElectronicEigenvalues.spin_channel.m_annotations = dict(
-    xml=MappingAnnotationModel(
-        path='calculation.eigenvalues.array.set.set[?"@comment"="spin 1"]'
-    )
-)
-ElectronicEigenvalues.reciprocal_cell.m_annotations = dict(
-    xml=MappingAnnotationModel(
-        path=ElectronicEigenvalues.spin_channel.m_annotations.xml
-        + '.set[?"@comment"="kpoint 1"]'
-    )  # TODO not going to work: add conversion to reference
+Simulation.m_annotations['xml'] = MappingAnnotationModel(path='modeling')
+
+
+Program.m_annotations['xml'] = MappingAnnotationModel(path='.generator')
+
+
+Program.name.m_annotations['xml'] = MappingAnnotationModel(
+    path='.i[?"@name"="program"]'
 )
-ElectronicEigenvalues.occupation.m_annotations = dict(
-    xml=MappingAnnotationModel(
-        path=ElectronicEigenvalues.reciprocal_cell.m_annotations.xml + '.r[0]'
-    )
+
+
+Program.version.m_annotations['xml'] = MappingAnnotationModel(
+    path='.i[?"@name"="version"]'
 )
-ElectronicEigenvalues.value.m_annotations = dict(
-    xml=MappingAnnotationModel(
-        path=ElectronicEigenvalues.reciprocal_cell.m_annotations.xml + '.r[1]'
-    )
+
+# ? compilation mode
+Program.compilation_host.m_annotations['xml'] = MappingAnnotationModel(
+    path='.i[?"@name"="platform"]'
 )
-# ? partial bands
 
 
 class VasprunXMLParser(MatchingParser):
@@ -128,19 +61,7 @@ def parse(
         logger: BoundLogger,
         child_archives: dict[str, EntryArchive] = None,
     ) -> None:
-        logger.info(
-            self.__class__.__repr__() + '.parse', parameter=configuration.parameter
-        )  # give feedback to Ahmed
-        archive_parser = MetainfoParser(
-            annotation_key='xml',
-            data_object=[
-                Simulation(
-                    model_system=[ModelSystem(cell=AtomicCell)],
-                    model_method=[DFT, KMesh],
-                    output=[ElectronicEigenvalues],
-                ),
-            ],
-        )
-        xml_parser = XMLParser(filepath='vasprun.xml')  # TODO apply match
-        xml_parser.convert(archive_parser)
-        archive.data = archive_parser.data_object
+        logger.info('VasprunXMLParser.parse', parameter=configuration.parameter)
+        data_parser = MetainfoParser(annotation_key='xml', data_object=Simulation())
+        XMLParser().convert(data_parser)
+        archive.data = data_parser.data_object