"""
This is the implementation of the exciting I/O functions
The functions are called with read write using the format "exciting"
"""
import numpy as np
import xml.etree.ElementTree as ET
from ase.atoms import Atoms
from ase.units import Bohr
from ase.utils import writer
from xml.dom import minidom
[docs]def read_exciting(fileobj, index=-1):
"""Reads structure from exiting xml file.
Parameters
----------
fileobj: file object
File handle from which data should be read.
Other parameters
----------------
index: integer -1
Not used in this implementation.
"""
# Parse file into element tree
doc = ET.parse(fileobj)
root = doc.getroot()
speciesnodes = root.find('structure').iter('species')
symbols = []
positions = []
basevects = []
atoms = None
# Collect data from tree
for speciesnode in speciesnodes:
symbol = speciesnode.get('speciesfile').split('.')[0]
natoms = speciesnode.iter('atom')
for atom in natoms:
x, y, z = atom.get('coord').split()
positions.append([float(x), float(y), float(z)])
symbols.append(symbol)
# scale unit cell accorting to scaling attributes
if 'scale' in doc.find('structure/crystal').attrib:
scale = float(str(doc.find('structure/crystal').attrib['scale']))
else:
scale = 1
if 'stretch' in doc.find('structure/crystal').attrib:
a, b, c = doc.find('structure/crystal').attrib['stretch'].text.split()
stretch = np.array([float(a), float(b), float(c)])
else:
stretch = np.array([1.0, 1.0, 1.0])
basevectsn = root.findall('structure/crystal/basevect')
for basevect in basevectsn:
x, y, z = basevect.text.split()
basevects.append(np.array([float(x) * Bohr * stretch[0],
float(y) * Bohr * stretch[1],
float(z) * Bohr * stretch[2]
]) * scale)
atoms = Atoms(symbols=symbols, cell=basevects)
atoms.set_scaled_positions(positions)
if 'molecule' in root.find('structure').attrib.keys():
if root.find('structure').attrib['molecule']:
atoms.set_pbc(False)
else:
atoms.set_pbc(True)
return atoms
[docs]@writer
def write_exciting(fileobj, images):
"""writes exciting input structure in XML
Parameters
----------
filename : str
Name of file to which data should be written.
images : Atom Object or List of Atoms objects
This function will write the first Atoms object to file.
Returns
-------
"""
root = atoms2etree(images)
rough_string = ET.tostring(root, 'utf-8')
reparsed = minidom.parseString(rough_string)
pretty = reparsed.toprettyxml(indent="\t")
fileobj.write(pretty)
def atoms2etree(images):
"""This function creates the XML DOM corresponding
to the structure for use in write and calculator
Parameters
----------
images : Atom Object or List of Atoms objects
Returns
-------
root : etree object
Element tree of exciting input file containing the structure
"""
if not isinstance(images, (list, tuple)):
images = [images]
root = ET.Element('input')
root.set(
'{http://www.w3.org/2001/XMLSchema-instance}noNamespaceSchemaLocation',
'http://xml.exciting-code.org/excitinginput.xsd')
title = ET.SubElement(root, 'title')
title.text = ''
structure = ET.SubElement(root, 'structure')
crystal = ET.SubElement(structure, 'crystal')
atoms = images[0]
for vec in atoms.cell:
basevect = ET.SubElement(crystal, 'basevect')
basevect.text = '%.14f %.14f %.14f' % tuple(vec / Bohr)
oldsymbol = ''
oldrmt = -1
newrmt = -1
scaled = atoms.get_scaled_positions()
for aindex, symbol in enumerate(atoms.get_chemical_symbols()):
if 'rmt' in atoms.arrays:
newrmt = atoms.get_array('rmt')[aindex] / Bohr
if symbol != oldsymbol or newrmt != oldrmt:
speciesnode = ET.SubElement(structure, 'species',
speciesfile='%s.xml' % symbol,
chemicalSymbol=symbol)
oldsymbol = symbol
if 'rmt' in atoms.arrays:
oldrmt = atoms.get_array('rmt')[aindex] / Bohr
if oldrmt > 0:
speciesnode.attrib['rmt'] = '%.4f' % oldrmt
atom = ET.SubElement(speciesnode, 'atom',
coord='%.14f %.14f %.14f' % tuple(scaled[aindex]))
if 'momenta' in atoms.arrays:
atom.attrib['bfcmt'] = '%.14f %.14f %.14f' % tuple(
atoms.get_array('mommenta')[aindex])
return root