Converted vaspparser to use metainfo

vasp @ 2c8b7763

-Subproject commit 8776a0bc7b32fb51e98ea8fe7af7d5630240edd3
+Subproject commit 2c8b7763eae8a27e7c1ca3a8ce86533978f42f73

nomad/cli/parse.py

@@ -48,8 +48,9 @@ def parse(
 
     if isinstance(parser_backend, MSection):
         backend = Backend(parser._metainfo_env, parser.domain)
-        root_section = str(datamodel.domains[parser.domain]['root_section'])
-        setattr(backend.entry_archive, root_section, parser_backend)
+        root_section = parser_backend.m_def.name
+        section_def = getattr(datamodel.EntryArchive, root_section)
+        backend.entry_archive.m_add_sub_section(section_def, parser_backend)
         backend.resource.add(parser_backend)
         parser_backend = backend
 

nomad/datamodel/metainfo/public.py

@@ -3439,6 +3439,11 @@ class section_run(MSection):
         repeats=True,
         a_legacy=LegacyDefinition(name='section_system'))
 
+    section_workflow = SubSection(
+        sub_section=SectionProxy('section_workflow'),
+        repeats=True
+    )
+
 
 class section_sampling_method(MSection):
     '''
@@ -5575,4 +5580,37 @@ class section_XC_functionals(MSection):
         a_legacy=LegacyDefinition(name='XC_functional_weight'))
 
 
+class section_workflow(MSection):
+    '''
+    Section containing the  results of a workflow.
+    '''
+
+    m_def = Section(validate=False, a_legacy=LegacyDefinition(name='section_workflow'))
+
+    workflow_type = Quantity(
+        type=str,
+        shape=[],
+        description='''
+        The type of calculation workflow. Can be one of relaxation, elastic, phonon,
+        molecular dynamics.
+        ''',
+        a_legacy=LegacyDefinition(name='workflow_type'))
+
+    relaxation_energy_tolerance = Quantity(
+        type=np.dtype(np.float64),
+        shape=[],
+        unit='joule',
+        description='''
+        The tolerance value in the energy between relaxation steps for convergence.
+        ''',
+        a_legacy=LegacyDefinition(name='relaxation_energy_tolerance'))
+
+    workflow_final_calculation_ref = Quantity(
+        type=Reference(SectionProxy('section_single_configuration_calculation')),
+        shape=[],
+        description='''
+        Reference to last calculation step.
+        ''')
+
+
 m_package.__init_metainfo__()

nomad/parsing/legacy.py

@@ -244,11 +244,14 @@ class Backend(AbstractParserBackend):
         if section_def.extends_base_section:
             section_def = section_def.base_sections[0]
 
+        section = self.__open_sections.get((section_def, -1), None)
+        if not section:
+            section = self.open_sections[(section_def, -1)]
         if isinstance(property_def, Quantity):
-            return self.__open_sections[(section_def, -1)].m_get(property_def)
+            return section.m_get(property_def)
 
         elif isinstance(property_def, SubSection):
-            return self.__open_sections[(section_def, -1)].m_get_sub_sections(property_def)
+            return section.m_get_sub_sections(property_def)
 
     def metaInfoEnv(self):
         if self.__legacy_env is None:
@@ -302,6 +305,12 @@ class Backend(AbstractParserBackend):
         section = self.resolve_context(context_uri)
         self.__close(section)
 
+    @property
+    def open_sections(self):
+        for section in self.entry_archive.m_all_contents():
+            self.__open(section)
+        return self.__open_sections
+
     def __open(self, section):
         if section.m_parent_index != -1:
             self.__open_sections[(section.m_def, section.m_parent_index)] = section
@@ -357,7 +366,9 @@ class Backend(AbstractParserBackend):
         if section_def.extends_base_section:
             section_def = section_def.base_sections[0]
 
-        section = self.__open_sections[(section_def, g_index)]
+        section = self.__open_sections.get((section_def, g_index), None)
+        if not section:
+            section = self.open_sections[(section_def, g_index)]
 
         return section, quantity_def
 

nomad/parsing/parsers.py

@@ -24,6 +24,7 @@ from .artificial import EmptyParser, GenerateRandomParser, TemplateParser, Chaos
 from eelsparser import EelsParser
 from mpesparser import MPESParser
 from aptfimparser import APTFIMParser
+from vaspparser import VASPParser
 
 try:
     # these packages are not available without parsing extra, which is ok, if the
@@ -121,18 +122,7 @@ parsers = [
         # mainfile_contents_re=r'',  # Empty regex since this code calls other DFT codes.
         mainfile_name_re=(r'.*/phonopy-FHI-aims-displacement-0*1/control.in$')
     ),
-    LegacyParser(
-        name='parsers/vasp', code_name='VASP', code_homepage='https://www.vasp.at/',
-        parser_class_name='vaspparser.VASPRunParser',
-        mainfile_mime_re=r'(application/.*)|(text/.*)',
-        mainfile_contents_re=(
-            r'^\s*<\?xml version="1\.0" encoding="ISO-8859-1"\?>\s*'
-            r'?\s*<modeling>'
-            r'?\s*<generator>'
-            r'?\s*<i name="program" type="string">\s*vasp\s*</i>'
-            r'?'),
-        supported_compressions=['gz', 'bz2', 'xz']
-    ),
+    VASPParser(),
     VaspOutcarParser(
         name='parsers/vasp-outcar', code_name='VASP', code_homepage='https://www.vasp.at/',
         parser_class_name='vaspparser.VaspOutcarParser',

nomad/processing/data.py

@@ -388,8 +388,9 @@ class Calc(Proc):
 
                 if isinstance(self._parser_backend, MSection):
                     backend = Backend(parser._metainfo_env, parser.domain)
-                    root_section = datamodel.domains[parser.domain]['root_section']
-                    setattr(backend.entry_archive, root_section, self._parser_backend)
+                    root_section = self._parser_backend.m_def.name
+                    section_def = getattr(datamodel.EntryArchive, root_section)
+                    backend.entry_archive.m_add_sub_section(section_def, self._parser_backend)
                     backend.resource.add(self._parser_backend)
                     self._parser_backend = backend
 

tests/test_parsing.py

@@ -289,13 +289,22 @@ def create_reference(data, pretty):
         return json.dumps(data, separators=(',', ':'))
 
 
-def assert_parser_result(backend, error=False):
-    status, errors = backend.status
-    assert status == 'ParseSuccess'
-    if error:
-        assert len(errors) > 0
-    else:
-        assert errors is None or len(errors) == 0
+@pytest.fixture(scope='function')
+def assert_parser_result(caplog):
+    def _assert(backend, error=False):
+        status, errors = backend.status
+        assert status == 'ParseSuccess'
+        if error:
+            if not errors:
+                errors = []
+                for record in caplog.get_records(when='call'):
+                    if record.levelname in ['WARNING', 'ERROR', 'CRITICAL']:
+                        errors.append(record.msg)
+            assert len(errors) > 0
+        else:
+            assert errors is None or len(errors) == 0
+
+    return _assert
 
 
 def assert_parser_dir_unchanged(previous_wd, current_wd):
@@ -308,8 +317,9 @@ def run_parser(parser_name, mainfile):
     result = parser.run(mainfile, logger=utils.get_logger(__name__))
     if isinstance(result, MSection):
         backend = Backend(parser._metainfo_env, parser.domain)
-        root_section = datamodel.domains[parser.domain]['root_section']
-        setattr(backend.entry_archive, root_section, result)
+        root_section = result.m_def.name
+        section_def = getattr(datamodel.EntryArchive, root_section)
+        backend.entry_archive.m_add_sub_section(section_def, result)
         backend.resource.add(result)
         result = backend
     result.domain = parser.domain
@@ -357,7 +367,7 @@ def add_calculation_info(backend: Backend, **kwargs) -> Backend:
 
 
 @pytest.mark.parametrize('parser_name, mainfile', parser_examples)
-def test_parser(parser_name, mainfile):
+def test_parser(parser_name, mainfile, assert_parser_result):
     previous_wd = os.getcwd()  # Get Working directory before parsing.
     parsed_example = run_parser(parser_name, mainfile)
     assert_parser_result(parsed_example)
@@ -365,7 +375,7 @@ def test_parser(parser_name, mainfile):
     assert_parser_dir_unchanged(previous_wd, current_wd=os.getcwd())
 
 
-def test_broken_xml_vasp():
+def test_broken_xml_vasp(assert_parser_result):
     parser_name, mainfile = 'parsers/vasp', 'tests/data/parsers/vasp/broken.xml'
     previous_wd = os.getcwd()  # Get Working directory before parsing.
     parsed_example = run_parser(parser_name, mainfile)