diff --git a/nomad/parsing/file_parser/file_parser.py b/nomad/parsing/file_parser/file_parser.py
index 7ecaf7511d46807185df539a495c67a783f2dbc2..a9dd56b443c5e55b1be6e99b55fd45c0d3bcaf40 100644
--- a/nomad/parsing/file_parser/file_parser.py
+++ b/nomad/parsing/file_parser/file_parser.py
@@ -20,6 +20,7 @@ import gzip
import bz2
import lzma
import tarfile
+from contextlib import contextmanager
from nomad.metainfo import MSection, SubSection
from nomad.utils import get_logger
@@ -96,6 +97,21 @@ class FileParser(ABC):
return self._mainfile_obj
+ @contextmanager
+ def open_mainfile_obj(self):
+ """
+ Returns the mainfile object with a context.
+ """
+ try:
+ self._mainfile_obj = self.open(self._mainfile)
+ yield self._mainfile_obj
+ except Exception:
+ pass
+ finally:
+ if self._mainfile_obj is not None:
+ self._mainfile_obj.close()
+ self._mainfile_obj = None
+
@property
def mainfile(self):
"""
@@ -138,7 +154,11 @@ class FileParser(ABC):
open_file = tarfile.open
else:
open_file = open
- return open_file(mainfile)
+
+ try:
+ return open_file(mainfile)
+ except Exception:
+ pass
def get(
self,
@@ -223,6 +243,21 @@ class FileParser(ABC):
string += f'--> {len(results)} parsed quantities ({", ".join(results[:5])}{", ..." if len(results) > 5 else ""})'
return string
+ def __enter__(self):
+ return self
+
+ def __exit__(self, exc_type, exc_val, exc_tb):
+ self.close()
+
+ def close(self):
+ if self._mainfile_obj:
+ self._mainfile_obj.close()
+ if self._file_handler is not None:
+ try:
+ self._file_handler.close()
+ except Exception:
+ pass
+
class Parser(ABC):
mainfile: str = None
diff --git a/nomad/parsing/file_parser/text_parser.py b/nomad/parsing/file_parser/text_parser.py
index af9d387c9dce8621ee6f00fb324a3c0e22cfc741..aa151d726523e7ec66c9bfc812ed845b68ec2faf 100644
--- a/nomad/parsing/file_parser/text_parser.py
+++ b/nomad/parsing/file_parser/text_parser.py
@@ -559,15 +559,17 @@ class TextParser(FileParser):
if quantity.sub_parser is not None:
self._parse_quantity(quantity)
- # free up memory
- self._file_handler = b' '
-
else:
for quantity in self._quantities:
if quantity.name == key or key is None:
if quantity.name not in self._results:
self._parse_quantity(quantity)
+ # free up memory
+ if isinstance(self._file_handler, mmap.mmap) and self.findall:
+ self._file_handler.close()
+ self._file_handler = b' '
+
return self
def clear(self):
@@ -603,7 +605,8 @@ class DataTextParser(TextParser):
data = np.loadtxt(self.mainfile)
else:
if not self._mainfile_contents and self.mainfile_obj:
- self._mainfile_contents = self.mainfile_obj.read()
+ with self.mainfile_obj as mainfile_obj:
+ self._mainfile_contents = mainfile_obj.read()
if self._mainfile_contents:
buffer = self._mainfile_contents
if isinstance(buffer, str):
diff --git a/nomad/parsing/file_parser/vasp_parser.py b/nomad/parsing/file_parser/vasp_parser.py
new file mode 100644
index 0000000000000000000000000000000000000000..adae8590087bdade00b55a0a8620b27a048afb4f
--- /dev/null
+++ b/nomad/parsing/file_parser/vasp_parser.py
@@ -0,0 +1,85 @@
+import numpy as np
+
+import runschema
+from nomad.parsing.file_parser.mapping_parser import MappingAnnotationModel, XMLParser
+
+
+class Program(runschema.run.Program):
+ version = runschema.run.Program.version
+ version.m_annotations = dict(
+ xml=MappingAnnotationModel(
+ # path=[".i[?_name=='version']"],
+ operator=(
+ 'get_version',
+ [
+ ".i[?_name=='version']",
+ ".i[?_name=='subversion']",
+ ".i[?_name=='platform']",
+ ],
+ ),
+ )
+ )
+
+ # compilation_datetime = runschema.run.Program.compilation_datetime
+ # compilation_datetime.m_annotations = dict(
+ # xml=XMLAnnotation(operator=[('get_compilation_datetime', ["i[?_name=='']"])])
+ # )
+
+
+class BandEnergies(runschema.calculation.BandEnergies):
+ n_spin_channels = runschema.calculation.BandEnergies.n_spin_channels
+ n_spin_channels.m_annotations = dict(
+ xml=MappingAnnotationModel(path='length(.array.set.set)')
+ )
+
+ n_kpoints = runschema.calculation.BandEnergies.n_kpoints
+ n_kpoints.m_annotations = dict(
+ xml=MappingAnnotationModel(path='length(.array.set.set[0].set)')
+ )
+
+ energies = runschema.calculation.BandEnergies.energies
+ energies.m_annotations = dict(
+ xml=MappingAnnotationModel(
+ operator=(
+ 'get_eigenvalues_energies',
+ [
+ '.array.set.set[].set[].r',
+ 'length(.array.set.set)',
+ 'length(.array.set.set[0].set)',
+ ],
+ ),
+ )
+ )
+
+
+class Calculation(runschema.calculation.Calculation):
+ eigenvalues = runschema.calculation.Calculation.eigenvalues
+ eigenvalues.m_annotations = dict(xml=MappingAnnotationModel(path='.eigenvalues'))
+
+
+class Run(runschema.run.Run):
+ program = runschema.run.Run.program
+ program.m_annotations = dict(xml=MappingAnnotationModel(path='.generator'))
+
+ calculation = runschema.run.Run.calculation
+ calculation.m_annotations = dict(xml=MappingAnnotationModel(path='.calculation'))
+
+
+runschema.run.Run.m_def.m_annotations = dict(
+ xml=MappingAnnotationModel(path='modeling')
+)
+
+
+class VASPXMLParser(XMLParser):
+ @staticmethod
+ def get_eigenvalues_energies(value, n_spin, n_kpoints):
+ array = np.transpose(value)[0].T
+ return np.reshape(array, (n_spin, n_kpoints, len(array[0])))
+
+ @staticmethod
+ def get_version(version, sub_version, platform):
+ return ' '.join([' '.join(s.split()) for s in [version, sub_version, platform]])
+
+ @staticmethod
+ def slice(value):
+ return np.array(value)[2:]
diff --git a/nomad/parsing/file_parser/xml_parser.py b/nomad/parsing/file_parser/xml_parser.py
index c2fd2d2958b5c505a556b6fe256f4cac61e48a6e..e1fecc6b3f32a359ad4c3ca96806cd56731ab4d8 100644
--- a/nomad/parsing/file_parser/xml_parser.py
+++ b/nomad/parsing/file_parser/xml_parser.py
@@ -58,21 +58,11 @@ class XMLParser(FileParser):
# I cannot use the lxml XMLParser directly because it is not compatible with
# the ElementTree implementation.
xml = etree.parse(
- self.mainfile_mainfile_obj, parser=etree.XMLParser(recover=True)
- )
- self._file_handler = ElementTree.fromstring(etree.tostring(xml))
- except Exception:
- pass
-
- self.logger.error('failed to load xml file')
- try:
- # I cannot use the lxml XMLParser directly because it is not compatible with
- # the ElementTree implementation.
- xml = etree.parse(
- self.open(self.mainfile), parser=etree.XMLParser(recover=True)
+ self.mainfile_obj, parser=etree.XMLParser(recover=True)
)
self._file_handler = ElementTree.fromstring(etree.tostring(xml))
except Exception:
+ self.logger.error('failed to load xml file')
pass
self.init_parameters()