Add orjson as required dependency as default JSON handler when custom encoder/decoder is not needed (#4411)

* clean up pyproject

* try to remove pybtex warning

* sort optional dependencies

* allow pyproject change to trigger test

* also include workflow and setup in test workflow trigger

* add orjson as dependency

* replace json.dumps

* replace json.loads

* fix phonondos numpy handling

* remove indent and see how many test will break

* replace json.load

* fix JSON str whitespace sensitivity

* replace json.dump

* fix os.path.exists

* replace str | PathLike

* use pmg typealias pathlike

* clean up import

* minor clean up of module docstring

Author: DanielYang59

.github/workflows/test.yml

@@ -5,10 +5,10 @@ name: test
 on:
   push:
     branches: [master]
-    paths: ["src/**/*.*", "tests/**/*.*"]
+    paths: ["src/**/*.*", "tests/**/*.*", ".github/workflows/*" ,"pyproject.toml", "setup.py"]
   pull_request:
     branches: [master]
-    paths: ["src/**/*.*", "tests/**/*.*"]
+    paths: ["src/**/*.*", "tests/**/*.*", ".github/workflows/*", "pyproject.toml", "setup.py" ]
   workflow_dispatch:
   workflow_call: # make this workflow reusable by release.yml
 

dev_scripts/chemenv/strategies/multi_weights_strategy_parameters.py

@@ -3,11 +3,11 @@
 from __future__ import annotations
 
 import copy
-import json
 from typing import TYPE_CHECKING
 
 import matplotlib.pyplot as plt
 import numpy as np
+import orjson
 
 from pymatgen.analysis.chemenv.coordination_environments.chemenv_strategies import (
     AngleNbSetWeight,
@@ -299,8 +299,8 @@ def get_weights(self, weights_options):
         "+-------------------------------------------------------------+\n"
     )
 
-    with open("ce_pairs.json", encoding="utf-8") as file:
-        ce_pairs = json.load(file)
+    with open("ce_pairs.json", "rb") as file:
+        ce_pairs = orjson.loads(file.read())
     self_weight_max_csms: dict[str, list[float]] = {}
     self_weight_max_csms_per_cn: dict[str, list[float]] = {}
     all_self_max_csms = []

docs/usage.md

@@ -55,14 +55,18 @@ classifiers = [
     "Topic :: Software Development :: Libraries :: Python Modules",
 ]
 dependencies = [
+    "bibtexparser>=1.4.0",
     "joblib>=1",
     "matplotlib>=3.8",
     "monty>=2025.1.9",
     "networkx>=2.7", # PR4116
+    # NumPy documentation suggests pinning the current major version as the C API is used
+    # https://numpy.org/devdocs/dev/depending_on_numpy.html#runtime-dependency-version-ranges
+    "numpy>=1.25.0,<3",
+    "orjson>=3.10,<4",
     "palettable>=3.3.3",
     "pandas>=2",
     "plotly>=5.0.0",
-    "bibtexparser>=1.4.0",
     "requests>=2.32",
     "ruamel.yaml>=0.17.0",
     "scipy>=1.13.0",
@@ -74,9 +78,6 @@ dependencies = [
     "tabulate>=0.9",
     "tqdm>=4.60",
     "uncertainties>=3.1.4",
-    # NumPy documentation suggests pinning the current major version as the C API is used
-    # https://numpy.org/devdocs/dev/depending_on_numpy.html#runtime-dependency-version-ranges
-    "numpy>=1.25.0,<3",
 ]
 version = "2025.5.2"
 
@@ -90,11 +91,19 @@ Pypi = "https://pypi.org/project/pymatgen"
 [project.optional-dependencies]
 abinit = ["netcdf4>=1.7.2"]
 ase = ["ase>=3.23.0"]
+ci = [
+    "pytest>=8.3.5",
+    "pytest-cov>=6.0.0",
+    "pytest-split>=0.10.0",
+]
 electronic_structure = ["fdint>=2.0.2"]
+matcalc = [
+    "matcalc>=0.4.0; python_version<'3.13'",
+    "matgl>=1.2.7; python_version<'3.13'",
+]
 mlp = ["matgl>=1.2.7 ; python_version<'3.13'"]
 numba = ["numba>=0.55"]
 numpy-v1 = ["numpy>=1.25.0,<2"] # Test NP1 on Windows (quite buggy ATM)
-zeopp = ["pyzeo"]  # Note: requires voro++ and zeo++ to be installed
 optional = [
     "pymatgen[abinit,ase,mlp,tblite,matcalc]",
     "beautifulsoup4",
@@ -106,7 +115,6 @@ optional = [
     "h5py>=3.11.0",
     "hiphive>=1.3.1",
     "jarvis-tools>=2020.7.14",
-    "matplotlib>=3.8",
     "phonopy>=2.33.3",
     "seekpath>=2.0.1",
 ]
@@ -117,15 +125,7 @@ symmetry = ["moyopy[interface]>=0.3", "spglib>=2.5"]
 # https://github.com/tblite/tblite/issues/175
 tblite = ["tblite[ase]>=0.3.0; platform_system=='Linux' and python_version<'3.12'"]
 vis = ["vtk>=6.0.0"]
-ci = [
-    "pytest>=8.3.5",
-    "pytest-cov>=6.0.0",
-    "pytest-split>=0.10.0",
-]
-matcalc = [
-    "matcalc>=0.4.0; python_version<'3.13'",
-    "matgl>=1.2.7; python_version<'3.13'",
-]
+zeopp = ["pyzeo"]  # Note: requires Voro++ and Zeo++ to be installed
 
 [project.scripts]
 pmg = "pymatgen.cli.pmg:main"
@@ -264,14 +264,8 @@ addopts = "--durations=30 --quiet -r xXs --color=yes --import-mode=importlib"
 filterwarnings = [
     # NOTE: the LAST matching option would be used
     "ignore::UserWarning", # Ignore UserWarning
-    "error:We strongly encourage explicit `encoding`:EncodingWarning", # Mark `zopen` EncodingWarning as error
-    "error:We strongly discourage using implicit binary/text:FutureWarning", # Mark `zopen` FutureWarning as error
-    # TODO: remove the following filter once `monty.io` dropped custom EncodingWarning
+    # TODO: remove the following filter once `monty.io` dropped custom EncodingWarning (python 3.10+ only)
     "error:We strongly encourage explicit `encoding`:monty.io.EncodingWarning",
-    # TODO: pybtex (perhaps some others) emits the following warnings
-    'ignore:pkg_resources is deprecated as an API:DeprecationWarning',
-    'ignore:distutils Version classes are deprecated:DeprecationWarning',
-    'ignore:Deprecated call to `pkg_resources.declare_namespace:DeprecationWarning',
 ]
 
 [tool.coverage.run]

pyproject.toml

@@ -35,6 +35,8 @@ numpy==2.0.0
     #   pandas
     #   scipy
     #   spglib
+orjson==3.10.18
+    # via pymatgen (pyproject.toml)
 packaging==24.2
     # via
     #   matplotlib

requirements.txt

@@ -11,6 +11,7 @@
 from typing import TYPE_CHECKING
 from warnings import warn
 
+import orjson
 from monty.json import MSONable, jsanitize
 
 from pymatgen.core.structure import Structure
@@ -193,7 +194,7 @@ def get_vasp_input(self, vasp_input_set: type[VaspInputSet] = MPRelaxSet, **kwar
             **kwargs: All keyword args supported by the VASP input set.
         """
         dct = vasp_input_set(self.final_structure, **kwargs).get_input_set()
-        dct["transformations.json"] = json.dumps(self.as_dict())
+        dct["transformations.json"] = orjson.dumps(self.as_dict()).decode()
         return dct
 
     def write_vasp_input(

src/pymatgen/alchemy/materials.py

@@ -11,11 +11,11 @@
 
 import abc
 import itertools
-import json
 import os
 from typing import TYPE_CHECKING
 
 import numpy as np
+import orjson
 from monty.json import MontyDecoder, MSONable
 from scipy.special import factorial
 
@@ -865,15 +865,15 @@ def __init__(self, permutations_safe_override=False, only_symbols=None):
                 data = file.readlines()
             for line in data:
                 cg_file = f"{MODULE_DIR}/{line.strip()}"
-                with open(cg_file, encoding="utf-8") as file:
-                    dd = json.load(file)
+                with open(cg_file, "rb") as file:
+                    dd = orjson.loads(file.read())
                 self.cg_list.append(CoordinationGeometry.from_dict(dd))
         else:
             for symbol in only_symbols:
                 fsymbol = symbol.replace(":", "#")
                 cg_file = f"{MODULE_DIR}/coordination_geometries_files/{fsymbol}.json"
-                with open(cg_file, encoding="utf-8") as file:
-                    dd = json.load(file)
+                with open(cg_file, "rb") as file:
+                    dd = orjson.loads(file.read())
                 self.cg_list.append(CoordinationGeometry.from_dict(dd))
 
         self.cg_list.append(CoordinationGeometry(UNKNOWN_ENVIRONMENT_SYMBOL, "Unknown environment", deactivate=True))

src/pymatgen/analysis/chemenv/coordination_environments/coordination_geometries.py

@@ -2,11 +2,11 @@
 
 from __future__ import annotations
 
-import json
-from os import makedirs
-from os.path import exists, expanduser
+import os
 from typing import TYPE_CHECKING
 
+import orjson
+
 from pymatgen.analysis.chemenv.utils.scripts_utils import strategies_class_lookup
 from pymatgen.core import SETTINGS
 
@@ -142,19 +142,19 @@ def save(self, root_dir=None):
             root_dir:
         """
         if root_dir is None:
-            home = expanduser("~")
+            home = os.expanduser("~")
             root_dir = f"{home}/.chemenv"
-        if not exists(root_dir):
-            makedirs(root_dir)
+        if not os.path.exists(root_dir):
+            os.makedirs(root_dir)
         config_dict = {"package_options": self.package_options}
         config_file = f"{root_dir}/config.json"
-        if exists(config_file):
+        if os.path.exists(config_file):
             test = input("Overwrite existing configuration ? (<Y> + <ENTER> to confirm)")
             if test != "Y":
                 print("Configuration not saved")
                 return config_file
-        with open(config_file, mode="w", encoding="utf-8") as file:
-            json.dump(config_dict, file)
+        with open(config_file, "wb") as file:
+            file.write(orjson.dumps(config_dict))
         print("Configuration saved")
         return config_file
 
@@ -167,12 +167,12 @@ def auto_load(cls, root_dir=None):
             root_dir:
         """
         if root_dir is None:
-            home = expanduser("~")
+            home = os.expanduser("~")
             root_dir = f"{home}/.chemenv"
         config_file = f"{root_dir}/config.json"
         try:
-            with open(config_file, encoding="utf-8") as file:
-                config_dict = json.load(file)
+            with open(config_file, "rb") as file:
+                config_dict = orjson.loads(file.read())
             return ChemEnvConfig(package_options=config_dict["package_options"])
 
         except OSError:

src/pymatgen/analysis/chemenv/utils/chemenv_config.py

@@ -22,13 +22,13 @@
 
 from __future__ import annotations
 
-import json
 import warnings
 from functools import lru_cache
 from itertools import groupby
 from typing import TYPE_CHECKING
 
 import numpy as np
+import orjson
 import plotly.express as px
 from monty.json import MSONable
 from plotly.graph_objects import Figure, Mesh3d, Scatter, Scatter3d
@@ -44,8 +44,8 @@
 if TYPE_CHECKING:
     from pymatgen.entries.computed_entries import ComputedEntry
 
-with open(f"{PKG_DIR}/util/plotly_chempot_layouts.json", encoding="utf-8") as file:
-    plotly_layouts = json.load(file)
+with open(f"{PKG_DIR}/util/plotly_chempot_layouts.json", "rb") as file:
+    plotly_layouts = orjson.loads(file.read())
 
 
 @due.dcite(

src/pymatgen/analysis/chempot_diagram.py

@@ -2,12 +2,12 @@
 
 from __future__ import annotations
 
-import json
 import math
 import os
 from typing import TYPE_CHECKING
 
 import numpy as np
+import orjson
 
 from pymatgen.analysis.diffraction.core import (
     AbstractDiffractionPatternCalculator,
@@ -29,9 +29,9 @@
 # This table was cited from "Neutron Data Booklet" 2nd ed (Old City 2003).
 with open(
     os.path.join(os.path.dirname(__file__), "neutron_scattering_length.json"),
-    encoding="utf-8",
+    "rb",
 ) as file:
-    ATOMIC_SCATTERING_LEN = json.load(file)
+    ATOMIC_SCATTERING_LEN = orjson.loads(file.read())
 
 
 class NDCalculator(AbstractDiffractionPatternCalculator):