Merge branch 'materialsproject:master' into master

.github/workflows/test.yml

@@ -46,7 +46,7 @@ jobs:
         run: conda install -c conda-forge openbabel
 
       - name: Install dependencies
-        run: uv pip install numpy==1.26.4 ; uv pip install -e '.[dev]' --system
+        run: uv pip install -e '.[dev]' --system
 
       - name: pytest
         env:

.pre-commit-config.yaml

@@ -8,29 +8,29 @@ ci:
 
 repos:
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.5.0
+    rev: v0.6.9
     hooks:
       - id: ruff
         args: [ --fix, --unsafe-fixes ]
       - id: ruff-format
 
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.6.0
+    rev: v5.0.0
     hooks:
       - id: check-yaml
       - id: end-of-file-fixer
       - id: trailing-whitespace
 
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.10.1
+    rev: v1.11.2
     hooks:
       - id: mypy
 
   - repo: https://github.com/codespell-project/codespell
     rev: v2.3.0
     hooks:
       - id: codespell
-        stages: [ commit, commit-msg ]
+        stages: [ commit-msg ]
         exclude_types: [ html ]
         additional_dependencies: [ tomli ] # needed to read pyproject.toml below py3.11
 
@@ -47,7 +47,7 @@ repos:
       - id: blacken-docs
 
   - repo: https://github.com/igorshubovych/markdownlint-cli
-    rev: v0.41.0
+    rev: v0.42.0
     hooks:
       - id: markdownlint
         # MD013: line too long
@@ -64,6 +64,6 @@ repos:
         args: [ --drop-empty-cells, --keep-output ]
 
   - repo: https://github.com/RobertCraigie/pyright-python
-    rev: v1.1.369
+    rev: v1.1.383
     hooks:
       - id: pyright

docs/changelog.md

@@ -6,6 +6,20 @@ nav_order: 2
 
 # Change Log
 
+## 2024.10.16
+* Add a update_incar option in VaspJob which updates parameters from a previous vasprun.xml.
+
+## 2024.10.15
+* Bug fix for pip installation.
+
+## 2024.10.14
+* PR #342 from @esoteric-ephemera (#342)
+    Minor update to the logic of the `auto_nbands` check for `VaspErrorHandler`. This check sees if the number of bands has been updated by VASP, and currently it only checks to see if that updated number is very large.
+    However, there are cases where the user specifies an NBANDS that is incompatible with their parallelization settings, as NBANDS must be divisible by $(\mathrm{ranks}) / (\mathrm{KPAR} \times \mathrm{NCORE})$. In these cases, VASP increases the number of bands to ensure the calculation can still proceed. This can happen in MP's band structure workflows with uniform $k$-point densities.
+    However, since the current `auto_nbands` handler applies no corrections to the job, these otherwise successful runs are killed.
+    This PR adds logic to ensure that the calculation is rerun with a higher number of bands appropriate to the parallelization setting. This is kinda redundant, since VASP already does this. But I think it has to occur this way because `VaspErrorHandler` is monitoring the job and flags it for an `auto_nbands` error.
+    Another implementation concern: it's generally safer to decrease the number of bands since this requires a lower energy cutoff to converge each band. It might be safer to decrease NBANDS as a fix
+
 ## 2024.6.24
 - Improved handling of ISMEAR for NKPT<4 (@Andrew-S-Rosen, @esoteric-ephemera)
 

docs/custodian.gaussian.handlers.md

@@ -8,7 +8,7 @@ nav_exclude: true
 
 This module implements error handlers for Gaussian runs.
 
-### *class* custodian.gaussian.handlers.GaussianErrorHandler(input_file: str, output_file: str, stderr_file: str = 'stderr.txt', cart_coords: bool = True, scf_max_cycles: int = 100, opt_max_cycles: int = 100, job_type: str = 'normal', lower_functional: str | None = None, lower_basis_set: str | None = None, prefix: str = 'error', check_convergence: bool = True)
+## *class* custodian.gaussian.handlers.GaussianErrorHandler(input_file: str, output_file: str, stderr_file: str = ‘stderr.txt’, cart_coords: bool = True, scf_max_cycles: int = 100, opt_max_cycles: int = 100, job_type: str = ‘normal’, lower_functional: str | None = None, lower_basis_set: str | None = None, prefix: str = ‘error’, check_convergence: bool = True)
 
 Bases: `ErrorHandler`
 
@@ -41,19 +41,19 @@ Initialize the GaussianErrorHandler class.
     be monitored and plotted (convergence criteria versus cycle number) and
     saved to a file called ‘convergence.png’.
 
-#### GRID_NAMES *= ('finegrid', 'fine', 'superfinegrid', 'superfine', 'coarsegrid', 'coarse', 'sg1grid', 'sg1', 'pass0grid', 'pass0')*
+### GRID_NAMES *= (‘finegrid’, ‘fine’, ‘superfinegrid’, ‘superfine’, ‘coarsegrid’, ‘coarse’, ‘sg1grid’, ‘sg1’, ‘pass0grid’, ‘pass0’)*
 
-#### MEM_UNITS *= ('kb', 'mb', 'gb', 'tb', 'kw', 'mw', 'gw', 'tw')*
+### MEM_UNITS *= (‘kb’, ‘mb’, ‘gb’, ‘tb’, ‘kw’, ‘mw’, ‘gw’, ‘tw’)*
 
-#### activate_better_guess *= False*
+### activate_better_guess *= False*
 
-#### check(directory: str = './')
+### check(directory: str = ‘./’)
 
 Check for errors in the Gaussian output file.
 
-#### conv_criteria*: ClassVar* *= {'max_disp': re.compile('\\\\s+(Maximum Displacement)\\\\s+(-?\\\\d+.?\\\\d\*|.\*)\\\\s+(-?\\\\d+.?\\\\d\*)'), 'max_force': re.compile('\\\\s+(Maximum Force)\\\\s+(-?\\\\d+.?\\\\d\*|.\*)\\\\s+(-?\\\\d+.?\\\\d\*)'), 'rms_disp': re.compile('\\\\s+(RMS {5}Displacement)\\\\s+(-?\\\\d+.?\\\\d\*|.\*)\\\\s+(-?\\\\d+.?\\\\d\*)'), 'rms_force': re.compile('\\\\s+(RMS {5}Force)\\\\s+(-?\\\\d+.?\\\\d\*|.\*)\\\\s+(-?\\\\d+.?\\\\d\*)')}*
+### conv_criteria\*: ClassVar\* *= {‘max_disp’: re.compile(’\\s+(Maximum Displacement)\\s+(-?\\d+.?\\d\*|.\*)\\s+(-?\\d+.?\\d\*)’), ‘max_force’: re.compile(’\\s+(Maximum Force)\\s+(-?\\d+.?\\d\*|.\*)\\s+(-?\\d+.?\\d\*)’), ‘rms_disp’: re.compile(’\\s+(RMS {5}Displacement)\\s+(-?\\d+.?\\d\*|.\*)\\s+(-?\\d+.?\\d\*)’), ‘rms_force’: re.compile(’\\s+(RMS {5}Force)\\s+(-?\\d+.?\\d\*|.\*)\\s+(-?\\d+.?\\d\*)’)}*
 
-#### *static* convert_mem(mem: float, unit: str)
+### *static* convert_mem(mem: float, unit: str)
 
 Convert memory size between different units to megabytes (MB).
 
@@ -67,19 +67,19 @@ Convert memory size between different units to megabytes (MB).
 * **Return type:**
   float
 
-#### correct(directory: str = './')
+### correct(directory: str = ‘./’)
 
 Perform necessary actions to correct the errors in the Gaussian output.
 
-#### error_defs*: ClassVar* *= {'A syntax error was detected in the input line.': 'syntax', 'Atom specifications unexpectedly found in input stream.': 'found_coords', 'Bad file opened by FileIO': 'bad_file', 'Convergence failure': 'scf_convergence', 'End of file in ZSymb': 'zmatrix', 'End of file reading connectivity.': 'coords', 'Error in internal coordinate system': 'internal_coords', 'FileIO operation on non-existent file.': 'missing_file', 'FormBX had a problem': 'linear_bend', 'Inv3 failed in PCMMkU': 'solute_solvent_surface', 'Linear angle in Tors.': 'linear_bend', 'No data on chk file.': 'empty_file', 'Optimization stopped': 'opt_steps', 'Out-of-memory error in routine': 'insufficient_mem', 'The combination of multiplicity ([0-9]+) and \\\\s+? ([0-9]+) electrons is impossible.': 'charge', 'There are no atoms in this input structure !': 'missing_mol', 'Z-matrix optimization but no Z-matrix variables.': 'coord_inputs'}*
+### error_defs\*: ClassVar\* *= {‘A syntax error was detected in the input line.’: ‘syntax’, ‘Atom specifications unexpectedly found in input stream.’: ‘found_coords’, ‘Bad file opened by FileIO’: ‘bad_file’, ‘Convergence failure’: ‘scf_convergence’, ‘End of file in ZSymb’: ‘zmatrix’, ‘End of file reading connectivity.’: ‘coords’, ‘Error in internal coordinate system’: ‘internal_coords’, ‘FileIO operation on non-existent file.’: ‘missing_file’, ‘FormBX had a problem’: ‘linear_bend’, ‘Inv3 failed in PCMMkU’: ‘solute_solvent_surface’, ‘Linear angle in Tors.’: ‘linear_bend’, ‘No data on chk file.’: ‘empty_file’, ‘Optimization stopped’: ‘opt_steps’, ‘Out-of-memory error in routine’: ‘insufficient_mem’, ‘The combination of multiplicity ([0-9]+) and \\s+? ([0-9]+) electrons is impossible.’: ‘charge’, ‘There are no atoms in this input structure !’: ‘missing_mol’, ‘Z-matrix optimization but no Z-matrix variables.’: ‘coord_inputs’}*
 
-#### error_patt *= re.compile('Optimization stopped|Convergence failure|FormBX had a problem|Linear angle in Tors.|Inv3 failed in PCMMkU|Error in internal coordinate system|End of file in ZSymb|There are no atoms in this input str)*
+### error_patt *= re.compile(‘Optimization stopped|Convergence failure|FormBX had a problem|Linear angle in Tors.|Inv3 failed in PCMMkU|Error in internal coordinate system|End of file in ZSymb|There are no atoms in this input str)*
 
-#### grid_patt *= re.compile('(-?\\\\d{5})')*
+### grid_patt *= re.compile(‘(-?\\d{5})’)*
 
-#### recom_mem_patt *= re.compile('Use %mem=([0-9]+)MW to provide the minimum amount of memory required to complete this step.')*
+### recom_mem_patt *= re.compile(‘Use %mem=([0-9]+)MW to provide the minimum amount of memory required to complete this step.’)*
 
-### *class* custodian.gaussian.handlers.WallTimeErrorHandler(wall_time: int, buffer_time: int, input_file: str, output_file: str, stderr_file: str = 'stderr.txt', prefix: str = 'error')
+## *class* custodian.gaussian.handlers.WallTimeErrorHandler(wall_time: int, buffer_time: int, input_file: str, output_file: str, stderr_file: str = ‘stderr.txt’, prefix: str = ‘error’)
 
 Bases: `ErrorHandler`
 
@@ -102,15 +102,15 @@ Initialize the WalTimeErrorHandler class.
     which means a series of error.1.tar.gz, error.2.tar.gz, … will be
     generated.
 
-#### check(directory: str = './')
+### check(directory: str = ‘./’)
 
 Check if the job is nearing the walltime. If so, return True, else False.
 
-#### correct(directory: str = './')
+### correct(directory: str = ‘./’)
 
 Perform the corrections.
 
-#### is_monitor*: bool* *= True*
+### is_monitor\*: bool\* *= True*
 
 This class property indicates whether the error handler is a monitor,
 i.e., a handler that monitors a job as it is running. If a

docs/custodian.gaussian.jobs.md

@@ -8,7 +8,7 @@ nav_exclude: true
 
 This module implements basic kinds of jobs for Gaussian runs.
 
-### *class* custodian.gaussian.jobs.GaussianJob(gaussian_cmd: str, input_file: str, output_file: str, stderr_file: str = 'stderr.txt', suffix: str = '', backup: bool = True)
+## *class* custodian.gaussian.jobs.GaussianJob(gaussian_cmd: str, input_file: str, output_file: str, stderr_file: str = ‘stderr.txt’, suffix: str = ‘’, backup: bool = True)
 
 Bases: `Job`
 
@@ -26,7 +26,7 @@ A basic Gaussian job.
   * **backup** (*bool*) – Whether to backup the input file. If True, the input will be
     copied with a “.orig” appended to the name. Defaults to True.
 
-#### *classmethod* generate_better_guess(gaussian_cmd: str, input_file: str, output_file: str, stderr_file: str = 'stderr.txt', backup: bool = True, cart_coords: bool = True, directory: str = './')
+### *classmethod* generate_better_guess(gaussian_cmd: str, input_file: str, output_file: str, stderr_file: str = ‘stderr.txt’, backup: bool = True, cart_coords: bool = True, directory: str = ‘./’)
 
 Generate a better initial guess for a Gaussian calculation (optimization or
 SCF run). This is done by running the job at a lower level of theory
@@ -46,15 +46,15 @@ generated by the previous job.
 * **Yields:**
   *GaussianJob* – The Gaussian job instance.
 
-#### postprocess(directory: str = './')
+### postprocess(directory: str = ‘./’)
 
 Perform any postprocessing of the Gaussian run. This includes making a copy
 of the input and output file if a suffix is specified.
 
 * **Parameters:**
   **directory** (*str*) – Directory where the job was run. Defaults to ‘./’.
 
-#### run(directory: str = './')
+### run(directory: str = ‘./’)
 
 Perform the actual Gaussian run.
 
@@ -65,15 +65,15 @@ Perform the actual Gaussian run.
 * **Return type:**
   subprocess.Popen
 
-#### setup(directory: str = './')
+### setup(directory: str = ‘./’)
 
 Perform initial setup for the job, i.e., make a backup of the input file if
 requested.
 
 * **Parameters:**
   **directory** (*str*) – Directory where the job will be run. Defaults to ‘./’.
 
-#### terminate(directory: str = './')
+### terminate(directory: str = ‘./’)
 
 Terminate the Gaussian job.
 

docs/custodian.gaussian.md

@@ -8,7 +8,6 @@ nav_exclude: true
 
 This package implements various Gaussian Jobs and Error Handlers.
 
-
 * [custodian.gaussian.handlers module](custodian.gaussian.handlers.md)
   * [`GaussianErrorHandler`](custodian.gaussian.handlers.md#custodian.gaussian.handlers.GaussianErrorHandler)
     * [`GaussianErrorHandler.GRID_NAMES`](custodian.gaussian.handlers.md#custodian.gaussian.handlers.GaussianErrorHandler.GRID_NAMES)

docs/custodian.vasp.utils.md

@@ -8,29 +8,30 @@ nav_exclude: true
 
 Utility methods for VASP error handlers.
 
-### custodian.vasp.utils.increase_k_point_density(kpoints: Kpoints | dict | float, structure: Structure, factor: float = 0.1, max_inc: int = 500, min_kpoints: int = 1, force_gamma: bool = True)
+## custodian.vasp.utils.increase_k_point_density(kpoints: Kpoints | dict | float, structure: Structure, factor: float = 0.1, max_inc: int = 500, min_kpoints: int = 1, force_gamma: bool = True)
 
 Inputs:
 : kpoints (Kpoints, dict, float, int) :
-  : If a dict or Kpoints object: original Kpoints used in the calculation
-    If a float: the original KSPACING used in the calculation
-  <br/>
-  structure (Structure) : associated structure
-  factor (float) : factor used to increase k-point density.
-  <br/>
-  > The first increase uses approximately (1 + factor) higher k-point density.
-  > The second increase: ~ (1 + 2\*factor) higher k-kpoint density, etc.
-  max_inc (int)
-  : before giving up
-  <br/>
-  min_kpoints (int): The minimum permitted number of k-points.
-  : Can be useful if using the tetrahedron method, where
-    at least 4 k-points are needed.
-  <br/>
-  force_gamma (bool) = True: whether to use Gamma-centered or
-  : Monkhorst-Pack grids
+: If a dict or Kpoints object: original Kpoints used in the calculation
+If a float: the original KSPACING used in the calculation
+<br/>
+structure (Structure) : associated structure
+factor (float) : factor used to increase k-point density.
+<br/>
+
+> The first increase uses approximately (1 + factor) higher k-point density.
+> The second increase: ~ (1 + 2\*factor) higher k-kpoint density, etc.
+> max_inc (int)
+> : before giving up
+> <br/>
+> min_kpoints (int): The minimum permitted number of k-points.
+> : Can be useful if using the tetrahedron method, where
+> at least 4 k-points are needed.
+> <br/>
+> force_gamma (bool) = True: whether to use Gamma-centered or
+> : Monkhorst-Pack grids
 
 Outputs:
 : dict :
-  : The new Kpoints object / KSPACING consistent with constraints.
-    If an empty dict, no new k-point mesh could be found.
+: The new Kpoints object / KSPACING consistent with constraints.
+If an empty dict, no new k-point mesh could be found.

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "custodian"
-version = "2024.6.24"
+version = "2024.10.16"
 description = "A simple JIT job management framework in Python."
 authors = [
     { name = "Janosh Riebesell", email = "janosh.riebesell@gmail.com" },
@@ -19,21 +19,20 @@ maintainers = [{ name = "Janosh Riebesell" }, { name = "Shyue Ping Ong" }]
 readme = "README.md"
 keywords = ["jit", "job", "just-in-time", "management", "vasp", "nwchem", "qchem"]
 classifiers = [
-    "Development Status :: 4 - Beta",
+    "Development Status :: 5 - Production/Stable",
     "Intended Audience :: Science/Research",
     "License :: OSI Approved :: MIT License",
     "Operating System :: OS Independent",
     "Programming Language :: Python :: 3 :: Only",
     "Programming Language :: Python :: 3",
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
-    "Programming Language :: Python :: 3.9",
     "Topic :: Scientific/Engineering :: Chemistry",
     "Topic :: Scientific/Engineering :: Physics",
     "Topic :: Software Development :: Libraries :: Python Modules",
 ]
 license = { file = "LICENSE" }
-requires-python = ">=3.9"
+requires-python = ">=3.10"
 
 dependencies = ["monty>=2.0.6", "psutil", "ruamel.yaml>=0.15.6"]
 
@@ -58,7 +57,7 @@ Package = "https://pypi.org/project/custodian"
 
 [tool.setuptools.packages.find]
 where = ["src"]
-include = ["custodian"]
+include = ["custodian*"]
 
 [tool.ruff]
 target-version = "py311"
@@ -163,3 +162,6 @@ reportMissingImports = false
 reportMissingModuleSource = false
 reportInvalidTypeForm = false
 exclude = ["**/tests"]
+
+[tool.setuptools.package-data]
+custodian = ["py.typed"]

src/custodian/vasp/jobs.py

@@ -84,6 +84,7 @@ def __init__(
         gamma_vasp_cmd=None,
         copy_magmom=False,
         auto_continue=False,
+        update_incar=False,
     ) -> None:
         """
         This constructor is necessarily complex due to the need for
@@ -136,6 +137,12 @@ def __init__(
                 if a STOPCAR is present. This is very useful if using the
                 wall-time handler which will write a read-only STOPCAR to
                 prevent VASP from deleting it once it finishes
+            update_incar (bool): Whether to update INCAR settings from updated settings
+                in vasprun.xml. This is particularly useful in certain calculatons where VASP
+                automatically sets certain parameters, e.g., NBANDS. Only parameters that are
+                already present in the INCAR will be updated, i.e., no new parameters will be
+                added even if they are in the final vasprun.xml. Note that settings_override take
+                precedence over updated params.
         """
         self.vasp_cmd = tuple(vasp_cmd)
         self.output_file = output_file
@@ -149,6 +156,7 @@ def __init__(
         self.gamma_vasp_cmd = tuple(gamma_vasp_cmd) if gamma_vasp_cmd else None
         self.copy_magmom = copy_magmom
         self.auto_continue = auto_continue
+        self.update_incar = update_incar
 
         if SENTRY_DSN:
             # if using Sentry logging, add specific VASP executable to scope
@@ -228,6 +236,17 @@ def setup(self, directory="./") -> None:
                     actions = self.auto_continue
                 dumpfn({"actions": actions}, os.path.join(directory, "continue.json"))
 
+        if self.update_incar:
+            try:
+                vasprun = Vasprun(os.path.join(directory, "vasprun.xml"))
+                params = vasprun.parameters
+                incar = Incar.from_file(os.path.join(directory, "INCAR"))
+                for k, v in incar.items():
+                    incar[k] = params.get(k, v)
+                incar.write_file(os.path.join(directory, "INCAR"))
+            except Exception as ex:
+                logger.error(f"Unable to update INCAR with params from vasprun.xml. {ex}")
+
         if self.settings_override is not None:
             VaspModder(directory=directory).apply_actions(self.settings_override)
 

tasks.py

@@ -8,10 +8,11 @@
 from glob import glob
 
 import requests
-from custodian import __version__ as CURRENT_VER
 from invoke import task
 from monty.os import cd
 
+from custodian import __version__ as CURRENT_VER
+
 NEW_VER = datetime.datetime.now(tz=datetime.UTC).strftime("%Y.%-m.%-d")
 
 

tests/ansible/test_interpreter.py

@@ -5,6 +5,7 @@
 from typing import Any
 
 import pytest
+
 from custodian.ansible.actions import FileActions
 from custodian.ansible.interpreter import Modder