Merge pull request #216 from datamol-io/new_rdkit

Compat with latest RDKit 2023.09

Author: hadim

.github/workflows/test.yml

@@ -16,15 +16,9 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["3.9", "3.10"]
+        python-version: ["3.10", "3.11"]
         os: ["ubuntu-latest", "macos-latest", "windows-latest"]
-        rdkit-version: ["2022.09", "2023.03"]
-
-        # just enable python 3.11 on ubuntu to not blow up the CI time.
-        include:
-          - os: ubuntu-latest
-            python-version: "3.11"
-            rdkit-version: "2023.03"
+        rdkit-version: ["2023.03", "2023.09"]
 
     runs-on: ${{ matrix.os }}
     timeout-minutes: 30

README.md

@@ -105,6 +105,7 @@ See below the associated versions of Python and RDKit, for which a minor version
 
 | `datamol` | `python`            | `rdkit`                       |
 | --------- | ------------------- | ----------------------------- |
+| `0.12.x`  | `[3.10, 3.11]`      | `[2023.03, 2023.09]`          |
 | `0.11.x`  | `[3.9, 3.10, 3.11]` | `[2022.09, 2023.03]`          |
 | `0.10.x`  | `[3.9, 3.10, 3.11]` | `[2022.03, 2022.09]`          |
 | `0.9.x`   | `[3.9, 3.10, 3.11]` | `[2022.03, 2022.09]`          |

datamol/cluster.py

@@ -242,7 +242,9 @@ def assign_to_centroids(
     features = dm.parallelized(feature_fn, all_mols, n_jobs=n_jobs)
 
     def distij(i, j, features=features):
-        return 1.0 - DataStructs.cDataStructs.TanimotoSimilarity(features[int(i)], features[int(j)])
+        return 1.0 - DataStructs.cDataStructs.TanimotoSimilarity(
+            features[int(i.item())], features[int(j.item())]
+        )
 
     if dist_fn is None:
         dist_fn = distij

datamol/mol.py

@@ -25,7 +25,6 @@
 from rdkit.Chem.Scaffolds import MurckoScaffold
 
 from rdkit.Chem.MolStandardize import rdMolStandardize
-from rdkit.Chem.MolStandardize import canonicalize_tautomer_smiles
 from rdkit.Chem.SaltRemover import SaltRemover
 
 import datamol
@@ -395,22 +394,19 @@ def sanitize_first(mols: List[Mol], charge_neutral: bool = False, sanifix: bool
     return None
 
 
-def standardize_smiles(smiles: str, tautomer: bool = False) -> str:
+def standardize_smiles(smiles: str) -> str:
     r"""
     Apply smile standardization procedure. This is a convenient function wrapped arrounf RDKit
     smiles standardizer and tautomeric canonicalization.
 
     Args:
         smiles: Smiles to standardize
-        tautomer: Whether to canonicalize tautomers
 
     Returns:
         standard_smiles: the standardized smiles
     """
 
     smiles = rdMolStandardize.StandardizeSmiles(smiles)
-    if tautomer:
-        smiles = canonicalize_tautomer_smiles(smiles)
     return smiles
 
 

docs/index.md

@@ -75,6 +75,7 @@ See below the associated versions of Python and RDKit, for which a minor version
 
 | `datamol` | `python`            | `rdkit`                       |
 | --------- | ------------------- | ----------------------------- |
+| `0.12.x`  | `[3.10, 3.11]`      | `[2023.03, 2023.09]`          |
 | `0.11.x`  | `[3.9, 3.10, 3.11]` | `[2022.09, 2023.03]`          |
 | `0.10.x`  | `[3.9, 3.10, 3.11]` | `[2022.03, 2022.09]`          |
 | `0.9.x`   | `[3.9, 3.10, 3.11]` | `[2022.03, 2022.09]`          |

pyproject.toml

@@ -83,9 +83,14 @@ minversion = "6.0"
 addopts = "--verbose --cov=datamol --cov-fail-under=85 --cov-report xml --cov-report term --durations=10 -n auto"
 testpaths = ["tests"]
 filterwarnings = [
-    "ignore::DeprecationWarning:rdkit.*:",
-    "ignore::DeprecationWarning:jupyter_client.*:",
-    "ignore::DeprecationWarning:pkg_resources.*:",
+    "ignore::DeprecationWarning:rdkit.Chem.MolStandardize",
+    "ignore::DeprecationWarning:jupyter_client",
+    "ignore::DeprecationWarning:pkg_resources",
+    "ignore::DeprecationWarning:joblib.externals.loky.backend",
+    "ignore::DeprecationWarning:dateutil.tz.tz",
+    "ignore::DeprecationWarning:joblib._utils",
+    "ignore::DeprecationWarning:openpyxl.packaging.core",
+    "ignore::DeprecationWarning:tqdm.std",
 ]
 
 [tool.coverage.run]

tests/test_mcs.py

@@ -12,11 +12,12 @@ def test_find_mcs():
     mols = [dm.to_mol(s) for s in smiles_list]
     smarts = dm.find_mcs(mols=mols, timeout=2)
 
-    # Load/export SMARTS to check RDKit versions compatibility.
-    excepted_smarts = "[#6&!R]-&!@[#6&!R]-&!@[#8&!R]-&!@[#6&R]1:&@[#6&R]:&@[#6&R]2:&@[#7&R]:&@[#6&R]:&@[#7&R]:&@[#6&R](:&@[#6&R]:&@2:&@[#6&R]:&@[#6&R]:&@1-&!@[#7&!R]-&!@[#6&!R](=&!@[#8&!R])-&!@[#6&!R]=&!@[#6&!R])-&!@[#7&!R]-&!@[#6&R]1:&@[#6&R]:&@[#6&R]:&@[#6&R]:&@[#6&R]:&@[#6&R]:&@1"
-    excepted_smarts_mol = dm.from_smarts(excepted_smarts)
-    excepted_smarts = dm.to_smarts(excepted_smarts_mol)
-
-    print(smarts)
+    # NOTE(hadim): hash are different given different RDKit version
+    expected_hashes = [
+        # RDKit >= 2023.09
+        "762f483ac10cc0f45c5aa2c790f9ef52f8dfb337",
+        # RDKit <= 2023.03
+        "49eff32e405d17980fad428cf4063ec52e2c5fda",
+    ]
 
-    assert smarts == excepted_smarts
+    assert dm.hash_mol(dm.from_smarts(smarts)) in expected_hashes

tests/test_mol.py

@@ -302,7 +302,7 @@ def test_sanitize_smiles_none():
 
 def test_standardize_smiles_tautomer():
     smiles = "C1=CC=CN=C1"
-    std_smiles = dm.standardize_smiles(smiles, tautomer=True)
+    std_smiles = dm.standardize_smiles(smiles)
     assert "c1ccncc1" == std_smiles
 
 

tests/test_utils_fs.py

@@ -17,7 +17,7 @@ def test_copy_files(tmp_path):
     dm.utils.fs.copy_file(source_path, destination_path)
 
     with open(destination_path) as f:
-        f.read() == content
+        assert f.read() == content
 
 
 def test_copy_dir(tmp_path):
@@ -49,10 +49,10 @@ def test_copy_dir(tmp_path):
     assert dm.utils.fs.is_file(file2_path)
 
     with open(file1_path) as f:
-        f.read() == content
+        assert f.read() == content
 
     with open(file2_path) as f:
-        f.read() == content
+        assert f.read() == content
 
 
 def test_mkdir(tmp_path):
@@ -91,7 +91,9 @@ def test_cache_dir():
 
 def test_get_mapper(tmp_path):
     fsmapper = dm.utils.fs.get_mapper(str(tmp_path / "test.txt"))
-    assert fsmapper.fs.protocol == "file"
+
+    # NOTE(hadim): depends the fsspec version
+    assert fsmapper.fs.protocol in ["file", ("file", "local")]
 
 
 @pytest.mark.skip_platform("win")