sparc.io module

Providing a new bundled SPARC file format

class sparc.io.SparcBundle(directory, mode='r', atoms=None, label=None, psp_dir=None, validator=<sparc.api.SparcAPI object>, cfg=<ase.config.Config object>)

Bases: object

Provide access to a calculation folder of SPARC as a simple bundle

The bundle can be optionally named as .sparc following the ASE’s .bundle format

Currently the write method only supports 1 image, while read method support reading atoms results in following conditions

1. No calculation (minimal): .ion + .inpt file –> 1 image

2) Single point calculation: .ion + .inpt + .out + .static –> 1 image with calc 3) Multiple SP calculations: chain all .out{digits} and .static{digitis} outputs 4) Relaxation: read from .geopt and .out (supporting chaining) 5) AIMD: read from .aimd and .out (support chaining)

directory

Path to the directory containing SPARC files.

Type:

Path

mode

File access mode (‘r’, ‘w’, or ‘a’).

Type:

str

label

Name of the main SPARC file.

Type:

str

init_atoms

Initial atomic configuration.

Type:

Atoms

init_inputs

Initial input parameters.

Type:

dict

psp_data

Pseudopotential data.

Type:

dict

raw_results

Raw results from SPARC calculations.

Type:

dict

psp_dir

Directory containing pseudopotentials.

Type:

Path

sorting

Sort order for atoms.

Type:

list

last_image

Index of the last image in a series of calculations.

Type:

int

validator

API validator for SPARC calculations.

Type:

SparcAPI

__find_psp_dir(psp_dir=None)

Finds the directory for SPARC pseudopotentials.

_find_files()

Finds all files matching the bundle label.

_make_label(label=None)

Infers or sets the label for the SPARC bundle.

_indir(ext, label=None, occur=0, d_format="{

02d}”): Finds a file with a specific extension in the bundle.

_read_ion_and_inpt()

Reads .ion and .inpt files together.

_write_ion_and_inpt()

Writes .ion and .inpt files to the bundle.

_read_results_from_index(index, d_format="{

02d}”): Reads results from a specific calculation index.

_make_singlepoint(calc_results, images, raw_results)

Converts results and images to SinglePointDFTCalculators.

_extract_static_results(raw_results, index="

“): Extracts results from static calculations.

_extract_geopt_results(raw_results, index="

“): Extracts results from geometric optimization calculations.

_extract_aimd_results(raw_results, index="

“): Extracts results from AIMD calculations.

convert_to_ase(index=-1, include_all_files=False, **kwargs)

Converts raw results to ASE Atoms with calculators.

read_raw_results(include_all_files=False)

Parses all files in the bundle and merges results.

read_psp_info()

Parses pseudopotential information from the inpt file.

convert_to_ase(index=-1, include_all_files=False, **kwargs)

Read the raw results from the bundle and create atoms with single point calculators

Parameters:

• index (int or str) – Index or slice of the image(s) to convert. Uses the same format as ase.io.read

• include_all_files (bool) – If true, also read results with indexed suffices

Returns:

ASE-atoms or images with single point results

Return type:

Atoms or List[Atoms]

psp_env = ['SPARC_PSP_PATH', 'SPARC_PP_PATH']

read_psp_info()

Parse the psp information from inpt file options The psp file locations are relative to the bundle.

If the files cannot be found, the dict will only contain the path

read_raw_results(include_all_files=False)

Parse all files using the given self.label. The results are merged dict from all file formats

Parameters:

include_all_files (bool) – Whether to include output files with different suffices If true: include all files (e.g. SPARC.out, SPARC.out_01, SPARC.out_02, etc).

Returns:

Dict containing all raw results. Only some of them will appear in the calculator’s results

Return type:

dict or List

Sets:

self.raw_results (dict or List): the same as the return value

#TODO: @TT 2024-11-01 allow accepting indices #TODO: @TT last_image is a bad name, it should refer to the occurance of images

the same goes with num_calculations

property resort

Wrap the self.sorting dict. If sorting information does not exist, use the default slicing

property sort

Wrap the self.sorting dict. If sorting information does not exist, use the default slicing

sparc.io.read_sparc(filename, index=-1, include_all_files=True, **kwargs)

Parse a SPARC bundle, return an Atoms object or list of Atoms (image) with embedded calculator result.

Parameters:

• filename (str or PosixPath) – Filename to the sparc bundle

• index (int or str) – Index or slice of the images, following the ase.io.read convention

• include_all_files (bool) – If true, parse all output files with indexed suffices

• **kwargs – Additional parameters

Returns:

Atoms or List[Atoms]

sparc.io.write_sparc(filename, images, **kwargs)

Write sparc file. Images can only be Atoms object or list of length 1

Parameters:

• filename (str or PosixPath) – Filename to the output sparc directory

• images (Atoms or List(Atoms)) – Atoms object to be written. Only supports writting 1 Atoms

• **kwargs – Additional parameters