Add raw parser of Lowdin charges from projwfc output file

aiida_quantumespresso/calculations/namelists.py

@@ -13,7 +13,6 @@
 from aiida.engine import CalcJob
 from aiida.orm import Dict
 from aiida.orm import RemoteData, FolderData, SinglefileData
-from aiida.plugins import ParserFactory
 
 from aiida_quantumespresso.calculations import _lowercase_dict, _uppercase_dict, _pop_parser_options
 from aiida_quantumespresso.utils.convert import convert_input_to_namelist_entry
@@ -39,7 +38,8 @@ class NamelistsCalculation(CalcJob):
     _default_namelists = ['INPUTPP']
     _blocked_keywords = []  # a list of tuples with key and value fixed
 
-    _retrieve_singlefile_list = []
+    _retrieve_temporary_list = ()
+    _retrieve_output_as_temp = False
 
     _DEFAULT_INPUT_FILE = 'aiida.in'
     _DEFAULT_OUTPUT_FILE = 'aiida.out'
@@ -174,12 +174,16 @@ def prepare_for_submission(self, folder):
 
         # Retrieve by default the output file and the xml file
         calcinfo.retrieve_list = []
-        calcinfo.retrieve_list.append(self.inputs.metadata.options.output_filename)
+        if not self._retrieve_output_as_temp:
+            calcinfo.retrieve_list.append(self.inputs.metadata.options.output_filename)
         settings_retrieve_list = settings.pop('ADDITIONAL_RETRIEVE_LIST', [])
         calcinfo.retrieve_list += settings_retrieve_list
         calcinfo.retrieve_list += self._internal_retrieve_list
 
-        calcinfo.retrieve_singlefile_list = self._retrieve_singlefile_list
+        calcinfo.retrieve_temporary_list = []
+        if self._retrieve_output_as_temp:
+            calcinfo.retrieve_temporary_list.append(self.inputs.metadata.options.output_filename)
+        calcinfo.retrieve_temporary_list += list(self._retrieve_temporary_list)
 
         # We might still have parser options in the settings dictionary: pop them.
         _pop_parser_options(self, settings)

aiida_quantumespresso/calculations/projwfc.py

@@ -3,6 +3,7 @@
 from aiida.orm import RemoteData, FolderData, Dict, XyData
 from aiida_quantumespresso.calculations.namelists import NamelistsCalculation
 
+
 class ProjwfcCalculation(NamelistsCalculation):
     """
     Projwfc.x code of the Quantum ESPRESSO distribution, handles the the
@@ -24,30 +25,41 @@ class ProjwfcCalculation(NamelistsCalculation):
     ]
     _default_parser = 'quantumespresso.projwfc'
     _internal_retrieve_list = [NamelistsCalculation._PREFIX + '.pdos*']
+    _retrieve_output_as_temp = True
 
     @classmethod
     def define(cls, spec):
         from aiida.orm import ProjectionData, BandsData
         super(ProjwfcCalculation, cls).define(spec)
         spec.input('parent_folder', valid_type=(RemoteData, FolderData), help='The output folder of a pw.x calculation')
         spec.output('output_parameters', valid_type=Dict)
+        spec.output('lowdin', valid_type=Dict)
         spec.output('Dos', valid_type=XyData)
         # if spin
-        spec.output('projections_up',   valid_type=ProjectionData, required=False)
+        spec.output('projections_up', valid_type=ProjectionData, required=False)
         spec.output('projections_down', valid_type=ProjectionData, required=False)
-        spec.output('bands_up',   valid_type=BandsData, required=False)
+        spec.output('bands_up', valid_type=BandsData, required=False)
         spec.output('bands_down', valid_type=BandsData, required=False)
         # if non-spin
         spec.output('projections', valid_type=ProjectionData, required=False)
         spec.output('bands', valid_type=BandsData, required=False)
         spec.default_output_node = 'output_parameters'
         spec.exit_code(
-            100, 'ERROR_NO_RETRIEVED_FOLDER', message='The retrieved folder data node could not be accessed.')
+            100, 'ERROR_NO_RETRIEVED_FOLDER', message='The retrieved folder data node could not be accessed.'
+        )
+        spec.exit_code(
+            101, 'ERROR_NO_RETRIEVED_TEMPORARY_FOLDER', message='The retrieved temporary folder could not be accessed.'
+        )
         spec.exit_code(
-            110, 'ERROR_READING_OUTPUT_FILE', message='The output file could not be read from the retrieved folder.')
+            110, 'ERROR_READING_OUTPUT_FILE', message='The output file could not be read from the retrieved folder.'
+        )
         spec.exit_code(
-            111, 'ERROR_READING_PDOSTOT_FILE', message='The pdos_tot file could not be read from the retrieved folder.')
+            111, 'ERROR_READING_PDOSTOT_FILE', message='The pdos_tot file could not be read from the retrieved folder.'
+        )
         spec.exit_code(
-            112, 'ERROR_PARSING_PROJECTIONS', message='An exception was raised parsing bands and projections.')
+            112, 'ERROR_PARSING_PROJECTIONS', message='An exception was raised parsing bands and projections.'
+        )
+        spec.exit_code(113, 'ERROR_PARSING_LOWDIN', message='An exception was raised parsing Lowdin charges.')
         spec.exit_code(
-            130, 'ERROR_JOB_NOT_DONE', message='The computation did not finish properly (\'JOB DONE\' not found).')
+            130, 'ERROR_JOB_NOT_DONE', message='The computation did not finish properly (\'JOB DONE\' not found).'
+        )

aiida_quantumespresso/parsers/parse_raw/projwfc.py

@@ -0,0 +1,58 @@
+from __future__ import absolute_import
+import re
+
+
+def parse_lowdin_charges(out_file_lines, lowdin_lines=None):
+    """Parse the Lowdin Charge section of the ``projwfc.x`` output.
+
+    :param out_file_lines: The projwfc.x stdout file content
+    :type out_file_lines: list[str]
+    :param lowdin_lines: indices of lines where 'Lowdin Charges:' has been found
+    :return: A tuple of lowdin data dict and spill parameter float
+    """
+    # find start of Lowdin charge output
+    if lowdin_lines is None:
+        lowdin_lines = []
+        for i, line in enumerate(out_file_lines):
+            if line.strip() == 'Lowdin Charges:':
+                lowdin_lines.append(i)
+
+    if len(lowdin_lines) > 1:
+        raise IOError("'Lowdin Charges:' found on multiple lines: {}".format(lowdin_lines))
+
+    if not lowdin_lines:
+        return None, None
+
+    spill_parameter = None
+    started_section = False
+    atom_index = None
+    output = {}
+    for i, line in enumerate(out_file_lines[lowdin_lines[0] + 1:]):
+        lineno = lowdin_lines[0] + 2 + i
+        # break on empty line or spill parameter
+        if not line.strip():
+            if started_section:
+                break
+            started_section = True
+            continue
+        if line.strip().startswith('Spilling Parameter'):
+            spill_parameter = float(line.strip().split()[-1])
+            break
+        atom_line_match = re.match(r'^\s*Atom\s\#\s*(\d+)\:(.*)$', line)
+        if atom_line_match:
+            atom_index = int(atom_line_match.group(1))
+            line = atom_line_match.group(2)
+        if atom_index is None:
+            raise IOError('No atom index specified on or before line {}'.format(lineno))
+        charge_type_match = re.match(r'^\s*(total charge|spin up|spin down|polarization)\s*\=\s*(\-?[\d\.]+)', line)
+        if charge_type_match:
+            charge_type = charge_type_match.group(1).replace(' ', '_')
+            output.setdefault(atom_index, {}).setdefault('sum', {})[charge_type] = float(charge_type_match.group(2))
+        else:
+            raise IOError('No charge type specified on line {}'.format(lineno))
+        for orbital_type, value in re.findall(
+            r'(s|p|d|f|px|py|pz|dxy|dxz|dyz|dz2|dx2-y2|f[xyz23\-\+]+)\s*\=\s*(\-?[\d\.]+)', line
+        ):
+            output.setdefault(atom_index, {}).setdefault(charge_type, {})[orbital_type] = float(value)
+
+    return output, spill_parameter