Merge pull request #222 from nlesc-nano/packaging

REL: Prepare for CAT 0.10.2

.github/workflows/pythonpackage.yml

@@ -53,7 +53,7 @@ jobs:
                 run: |
                     case "${{ matrix.special }}" in
                         "; minimum version")
-                            conda create -n test -c conda-forge python=${{ matrix.version }} rdkit==2018.03.1 numpy=1.15.0 pandas=0.23.0 h5py=2.7.0 scipy=0.19.1 contextlib2==0.6.0 typing-extensions==3.7.4.3 pytest==5.4.0 pyyaml=5.1.0 pytest-cov pytest-mock
+                            conda create -n test -c conda-forge python=${{ matrix.version }} rdkit==2018.03.1 numpy=1.15.0 pandas=0.23.0 h5py=2.7.0 scipy=0.19.1 contextlib2==0.6.0 typing-extensions==3.7.4.3 pytest==5.4.0 pyyaml=5.1.0 packaging==16.8 pytest-cov pytest-mock
                             source $CONDA/bin/activate test
                             pip install schema==0.7.2 Nano-Utils==0.4.3 AssertionLib==2.2.3 plams==1.5.1 qmflows==0.11.0
                             pip install -e . --no-dependencies

CAT/_mol_str_parser.py

@@ -20,18 +20,19 @@
 import textwrap
 from typing import Any, Callable
 
-from nanoutils import PartialPrepend, VersionInfo
+from nanoutils import PartialPrepend
+from packaging.version import Version
 
 try:
     import rdkit
     from rdkit import Chem
 except ModuleNotFoundError:
     # Precaution against sphinx mocking raising a TypeError
     FLAGS = 0
-    RDKIT_VERSION = VersionInfo(0, 0, 0)
+    RDKIT_VERSION = Version("0.0.0")
 else:
     FLAGS = Chem.SanitizeFlags.SANITIZE_ALL ^ Chem.SanitizeFlags.SANITIZE_ADJUSTHS
-    RDKIT_VERSION = VersionInfo.from_str(rdkit.__version__)
+    RDKIT_VERSION = Version(rdkit.__version__)
 
 __all__ = ["FormatEnum", "str_to_rdmol"]
 
@@ -72,8 +73,8 @@ class FormatEnum(enum.Enum):
     SMILES = PartialPrepend(str_to_rdmol, Chem.MolFromSmiles, "SMILES string", sanitize=False)
     TPL = PartialPrepend(str_to_rdmol, Chem.MolFromTPLBlock, "TPL block", sanitize=False)
     TPL_FILE = PartialPrepend(str_to_rdmol, Chem.MolFromTPLFile, "TPL file", sanitize=False)
-    if RDKIT_VERSION >= (2020, 3, 3):
+    if RDKIT_VERSION >= Version("2020.03.3"):
         SVG = PartialPrepend(str_to_rdmol, Chem.MolFromRDKitSVG, "SVG string", sanitize=False, removeHs=False)
-    if RDKIT_VERSION >= (2020, 9, 1):
+    if RDKIT_VERSION >= Version("2020.09.1"):
         PNG = PartialPrepend(str_to_rdmol, Chem.MolFromPNGString, "PNG string")
         PNG_FILE = PartialPrepend(str_to_rdmol, Chem.MolFromPNGFile, "PNG file")

CHANGELOG.rst

@@ -8,7 +8,11 @@ This project adheres to `Semantic Versioning <http://semver.org/>`_.
 
 0.10.2
 ******
-* *placeholder*.
+* Added the option to specify dummy atoms as anchor group.
+* Added the option to specify the format type of anchor groups (SMILES, SMARTS, *etc.*).
+* Fixed an issue wherein anchor-group-parsing could fail for mono-atomic anchors.
+* Added ``packaging`` and ``rdkit-pypi`` as dependencies;
+  manually installing ``rdkit`` via conda is no longer necessary.
 
 
 0.10.1

conftest.py

@@ -6,16 +6,15 @@
 
 import pytest
 import rdkit
-from nanoutils import VersionInfo
+from packaging.version import Version
 
 from CAT import logger
 
 
 @pytest.fixture(scope="session", autouse=True)
 def rdkit_version() -> None:
     """Validate the rdkit version."""
-    rdkit_version = VersionInfo.from_str(rdkit.__version__)
-    if rdkit_version < (2021, 3, 4):
+    if Version(rdkit.__version__) < Version("2021.03.4"):
         warnings.warn(
             "The CAT test suite requires RDKit >= 2021.03.4 to properly run; "
             f"observed version: {rdkit.__version__}"

setup.py

@@ -19,7 +19,7 @@
     version=version['__version__'],
     description=('A collection of tools designed for the automatic '
                  'construction of chemical compounds.'),
-    long_description=f'{readme}\n\n',
+    long_description=readme + "\n\n",
     long_description_content_type='text/x-rst',
     author=['Bas van Beek', 'Jelena Belic'],
     author_email='[email protected]',
@@ -100,6 +100,7 @@
         'typing-extensions>=3.7.4.3',
         'qmflows>=0.11.0',
         'rdkit-pypi>=2018.03.1',
+        'packaging>=1.16.8',
     ],
     setup_requires=[
         'pytest-runner',

tests/test_entry_points.py

@@ -11,7 +11,7 @@
 import pytest
 import numpy as np
 from scm.plams import Molecule
-from nanoutils import VersionInfo
+from packaging.version import Version
 
 from CAT.data_handling.entry_points import main
 
@@ -27,8 +27,6 @@
 DB_PATH = PATH / 'database'
 PATH_REF = PATH / 'ligand_ref'
 
-RDKIT_VERSION = VersionInfo.from_str(rdkit.__version__)
-
 
 class TestMain:
     @pytest.fixture(scope="class", autouse=True)
@@ -60,7 +58,10 @@ def test_raise(self) -> None:
             main([f'{filename}bob'])
 
     @pytest.mark.parametrize("filename", PATH_DICT.values(), ids=PATH_DICT.keys())
-    @pytest.mark.xfail(RDKIT_VERSION < (2021, 3, 4), reason="requires rdkit >= 2021.03.4")
+    @pytest.mark.xfail(
+        Version(rdkit.__version__) < Version("2021.03.4"),
+        reason="requires rdkit >= 2021.03.4",
+    )
     def test_mol(self, filename: str) -> None:
         mol = Molecule(LIG_PATH / filename)
         mol_ref = Molecule(PATH_REF / filename)

tests/test_gen_job_manager.py

@@ -1,14 +1,13 @@
 """Tests for the :mod:`CAT.gen_job_manager.GenJobManager` class in :mod:`CAT.gen_job_manager`."""
 
-import re
 from os.path import (join, abspath)
 from collections import abc
 
 import pytest
 import numpy as np
 from scm.plams import Settings, JobManager, AMSJob, AMSResults
 from assertionlib import assertion
-from nanoutils import VersionInfo
+from packaging.version import Version
 
 from CAT.gen_job_manager import GenJobManager
 from CAT.utils import get_formula
@@ -19,11 +18,6 @@
 FOLDER = 'test_plams_workdir'
 
 
-NUMPY_VERSION = VersionInfo._make(
-    int(i) for i in re.match(r'(\d+).(\d+).(\d+)', np.__version__).groups()
-)
-
-
 def test_init() -> None:
     """Test :meth:`CAT.gen_job_manager.GenJobManager.__init__`."""
     manager = GenJobManager(SETTINGS, PATH, FOLDER)
@@ -40,7 +34,10 @@ def test_init() -> None:
     assertion.eq(manager.input, join(ABSPATH, FOLDER, SETTINGS.hashing))
 
 
-@pytest.mark.skipif(NUMPY_VERSION < (1, 16, 0), reason="Unpickling requires numpy >= 1.16.0")
+@pytest.mark.skipif(
+    Version(np.__version__) < Version("1.16.0"),
+    reason="Unpickling requires numpy >= 1.16.0",
+)
 def test_load_job() -> None:
     """Test :meth:`CAT.gen_job_manager.GenJobManager.load_job`."""
     filename = join(PATH, FOLDER, 'QD_opt_part1', 'QD_opt_part1.hash')

tests/test_ligand_anchoring.py

@@ -19,7 +19,7 @@
 from scm.plams import from_smiles, Molecule, to_rdmol, PT
 from assertionlib import assertion
 from schema import SchemaError
-from nanoutils import VersionInfo
+from packaging.version import Version
 
 from CAT.utils import get_template, KindEnum, AnchorTup, FormatEnum
 from CAT.base import prep_input
@@ -36,8 +36,6 @@
 
 PATH = Path('tests') / 'test_files'
 
-RDKIT_VERSION = VersionInfo.from_str(rdkit.__version__)
-
 
 class TestGetFunctionalGroups:
     """Tests for :func:`CAT.attachment.ligand_anchoring.get_functional_groups`."""
@@ -158,7 +156,10 @@ def get_h5py_group(self) -> Generator[h5py.Group, None, None]:
     })
 
     @pytest.mark.parametrize("kwargs_id,kwargs", OPTIONS_DICT.items(), ids=OPTIONS_DICT.keys())
-    @pytest.mark.xfail(RDKIT_VERSION < (2021, 3, 4), reason="requires rdkit >= 2021.03.4")
+    @pytest.mark.xfail(
+        Version(rdkit.__version__) < Version("2021.03.4"),
+        reason="requires rdkit >= 2021.03.4",
+    )
     def test_options(self, kwargs_id: str, kwargs: Dict[str, Any], group: h5py.Group) -> None:
         # Parse the anchoring options
         mol = from_smiles("OC(=O)CCCCC")

tests/test_ligand_attach.py

@@ -10,7 +10,7 @@
 import h5py
 import numpy as np
 from assertionlib import assertion
-from nanoutils import VersionInfo
+from packaging.version import Version
 from scm.plams import Settings, Molecule
 
 from CAT.base import prep
@@ -26,8 +26,6 @@
 QD_PATH = PATH / 'qd'
 DB_PATH = PATH / 'database'
 
-RDKIT_VERSION = VersionInfo.from_str(rdkit.__version__)
-
 
 def test_get_rotmat1() -> None:
     """Test :func:`CAT.attachment.ligand_attach._get_rotmat1`."""
@@ -119,7 +117,10 @@ def run_cat(self, request: "_pytest.fixtures.SubRequest") -> Generator[DihedTup,
         for file in files:
             shutil.rmtree(file, ignore_errors=True)
 
-    @pytest.mark.xfail(RDKIT_VERSION < (2021, 3, 4), reason="requires rdkit >= 2021.03.4")
+    @pytest.mark.xfail(
+        Version(rdkit.__version__) < Version("2021.03.4"),
+        reason="requires rdkit >= 2021.03.4",
+    )
     def test_atoms(self, output: DihedTup) -> None:
         dtype = [("symbols", "U2"), ("coords", "f8", 3)]
         atoms = np.fromiter(
@@ -159,7 +160,10 @@ def run_cat(
         for file in files:
             shutil.rmtree(file, ignore_errors=True)
 
-    @pytest.mark.xfail(RDKIT_VERSION < (2021, 3, 4), reason="requires rdkit >= 2021.03.4")
+    @pytest.mark.xfail(
+        Version(rdkit.__version__) < Version("2021.03.4"),
+        reason="requires rdkit >= 2021.03.4",
+    )
     def test_atoms(self, output: AllignmentTup) -> None:
         dtype = [("symbols", "S2"), ("coords", "f8", 3)]
         iterator = ((at.symbol, at.coords) for at in output.mol)
@@ -215,7 +219,10 @@ def run_cat(
         for file in files:
             shutil.rmtree(file, ignore_errors=True)
 
-    @pytest.mark.xfail(RDKIT_VERSION < (2021, 3, 4), reason="requires rdkit >= 2021.03.4")
+    @pytest.mark.xfail(
+        Version(rdkit.__version__) < Version("2021.03.4"),
+        reason="requires rdkit >= 2021.03.4",
+    )
     def test_atoms(self, output: AllignmentTup) -> None:
         dtype = [("symbols", "S2"), ("coords", "f8", 3)]
         iterator = ((at.symbol, at.coords) for at in output.mol)

tests/test_multi_ligand.py

@@ -8,8 +8,9 @@
 
 import pytest
 import rdkit
-from nanoutils import delete_finally, VersionInfo
+from nanoutils import delete_finally
 from scm.plams import Settings
+from packaging.version import Version
 from CAT import base
 
 PATH = Path('tests') / 'test_files'
@@ -18,8 +19,6 @@
 QD_PATH = PATH / 'qd'
 DB_PATH = PATH / 'database'
 
-RDKIT_VERSION = VersionInfo.from_str(rdkit.__version__)
-
 with open(PATH / 'input3.yaml', 'r') as f:
     SETTINGS = Settings(yaml.load(f, Loader=yaml.FullLoader))
 
@@ -33,7 +32,10 @@ def _iterfiles(file1, file2):
 
 
 @delete_finally(LIG_PATH, QD_PATH, DB_PATH)
-@pytest.mark.xfail(RDKIT_VERSION < (2021, 3, 4), reason="requires rdkit >= 2021.03.4")
+@pytest.mark.xfail(
+    Version(rdkit.__version__) < Version("2021.03.4"),
+    reason="requires rdkit >= 2021.03.4",
+)
 def test_multi_ligand() -> None:
     """Test :mod:`CAT.multi_ligand`."""
     base.prep(SETTINGS.copy())