address comments and linting

package/CHANGELOG

@@ -16,7 +16,7 @@ The rules for this file:
 -------------------------------------------------------------------------------
 ??/??/?? IAlibay, HeetVekariya, marinegor, lilyminium, RMeli,
          ljwoods2, aditya292002, pstaerk, PicoCentauri, BFedder,
-         tyler.je.reddy, SampurnaM, leonwehrhan, kainszs
+         tyler.je.reddy, SampurnaM, leonwehrhan, kainszs, cbouy
 
  * 2.8.0
 
@@ -36,6 +36,8 @@ Fixes
  * Fix groups.py doctests using sphinx directives (Issue #3925, PR #4374)
 
 Enhancements
+ * Added `TemplateInferer` and `RDKitInferer` dataclasses to the
+   `RDKitInferring` module to be used by the RDKit converter. (PR #4305)
  * Added a tqdm progress bar for `MDAnalysis.analysis.pca.PCA.transform()`
    (PR #4531)
  * Improved performance of PDBWriter (Issue #2785, PR #4472)
@@ -46,6 +48,9 @@ Enhancements
    DOI 10.1021/acs.jpcb.7b11988. (Issue #2039, PR #4524)
 
 Changes
+ * Refactored the RDKit converter code to move the inferring code in a separate
+   `RDKitInferring` module. The bond order and charges inferer has been move to
+   an `MDAnalysisInferer` dataclass in there. (PR #4305)
  * As per SPEC0 the minimum supported Python version has been raised
    to 3.10 (PR #4502)
  * MDAnalysis.analysis.hole2 is now directly imported from the mdakit
@@ -63,6 +68,8 @@ Changes
    numpy.testing.assert_allclose #4438)
 
 Deprecations
+ * The RDKit converter parameter `NoImplicit` has been deprecated in favour of
+   `implicit_hydrogens` and `inferer` parameters. (PR #4305)
  * The MDAnalysis.analysis.waterdynamics module has been deprecated in favour
    of the waterdynamics MDAKit and will be removed in version 3.0.0 (PR #4404)
  * The MDAnalysis.analysis.psa module has been deprecated in favour of

package/MDAnalysis/converters/RDKit.py

@@ -257,6 +257,9 @@ class RDKitConverter(base.ConverterBase):
         >>> mol = u.atoms.convert_to.rdkit(inferer=infer_from_template)
 
 
+    Note that all builtin inferer functions can be found in the
+    :mod:`RDKitInferring` module.
+
     Since one of the main use case of the converter is converting trajectories
     and not just a topology, creating a new molecule from scratch for every
     frame would be too slow so the converter uses a caching system. The cache
@@ -292,7 +295,7 @@ class RDKitConverter(base.ConverterBase):
         files while the original name is still available through
         ``atom.GetProp("_MDAnalysis_name")``
 
-    .. versionchanged:: 2.7.0
+    .. versionchanged:: 2.8.0
         Deprecated ``max_iter`` (moved to the inferer class
         :class:`~MDAnalysis.converters.RDKitInferring.MDAnalysisInferer`) and
         deprecated ``NoImplicit`` in favor of ``implicit_hydrogens``. Added
@@ -341,20 +344,20 @@ def convert(
             raise ImportError(
                 "RDKit is required for the RDKitConverter but "
                 "it's not installed. Try installing it with \n"
-                "conda install -c conda-forge rdkit"
-            )
+                "`conda install -c conda-forge rdkit` or `pip install rdkit`"
+            ) from None
         try:
             # make sure to use atoms (Issue 46)
             ag = obj.atoms
         except AttributeError:
             raise TypeError(
-                "No `atoms` attribute in object of type {}, "
-                "please use a valid AtomGroup or Universe".format(type(obj))
+                f"No `atoms` attribute in object of type {type(obj)!r}, "
+                "please use a valid AtomGroup or Universe"
             ) from None
 
         if (max_iter := kwargs.get("max_iter")) is not None:
             warnings.warn(
-                "Using `max_iter` is deprecated, use `MDAnalysisInferer "
+                "Using `max_iter` is deprecated, use `MDAnalysisInferer"
                 "(max_iter=...)` instead",
                 DeprecationWarning,
             )
@@ -427,6 +430,12 @@ def atomgroup_to_mol(
     inferer : Optional[Callable[[rdkit.Chem.rdchem.Mol], rdkit.Chem.rdchem.Mol]]
         A callable to infer bond orders and charges for the RDKit molecule
         created by the converter. If ``None``, inferring is skipped.
+
+
+    .. versionchanged:: 2.8.0
+        Deprecated ``NoImplicit`` in favor of ``implicit_hydrogens``. Added
+        ``inferer`` to specify a callable that can transform the molecule (this
+        operation is cached).
     """
     try:
         elements = ag.elements
@@ -469,7 +478,7 @@ def atomgroup_to_mol(
 
     # attributes accepted in PDBResidueInfo object
     pdb_attrs = {}
-    for attr in RDATTRIBUTES.keys():
+    for attr in RDATTRIBUTES:
         if hasattr(ag, attr):
             pdb_attrs[attr] = getattr(ag, attr)
     resnames = pdb_attrs.get("resnames", None)
@@ -504,10 +513,9 @@ def get_resname(idx):
             _add_mda_attr_to_rdkit(attr, values[i], mi, get_resname(i))
         rdatom.SetMonomerInfo(mi)
         # other properties
-        for attr in other_attrs.keys():
-            value = other_attrs[attr][i]
-            attr = "_MDAnalysis_%s" % _TOPOLOGY_ATTRS[attr].singular
-            _set_atom_property(rdatom, attr, value)
+        for attr, values in other_attrs.items():
+            attr = f"_MDAnalysis_{_TOPOLOGY_ATTRS[attr].singular}"
+            _set_atom_property(rdatom, attr, values[i])
         _set_atom_property(rdatom, "_MDAnalysis_index", i)
         # add atom
         index = mol.AddAtom(rdatom)
@@ -577,7 +585,7 @@ def _add_mda_attr_to_rdkit(attr, value, mi, resname=""):
 
     # set attribute value in RDKit MonomerInfo
     rdattr = RDATTRIBUTES[attr]
-    getattr(mi, "Set%s" % rdattr)(value)
+    getattr(mi, f"Set{rdattr}")(value)
 
 
 def _set_str_prop(atom, attr, value):

package/MDAnalysis/converters/RDKitInferring.py

@@ -5,7 +5,7 @@
 # Copyright (c) 2006-2017 The MDAnalysis Development Team and contributors
 # (see the file AUTHORS for the full list of names)
 #
-# Released under the GNU Public Licence, v2 or any higher version
+# Released under the GNU Lesser General Public License, v2 or higher
 #
 # Please cite your use of MDAnalysis in published work:
 #
@@ -69,8 +69,8 @@
 
 with suppress(ImportError):
     from rdkit.Chem.rdDetermineBonds import (
-        DetermineBondOrders,  # available since 2022.09.1
-    )
+        DetermineBondOrders,
+    )  # available since 2022.09.1
 
 
 def reorder_atoms(
@@ -124,7 +124,7 @@ class MDAnalysisInferer:
         fix challenging cases must have single reactant and product, and
         cannot add any atom.
 
-    .. versionadded:: 2.7.0
+    .. versionadded:: 2.8.0
     """
 
     MONATOMIC_CATION_CHARGES: ClassVar[Dict[int, int]] = {
@@ -624,6 +624,8 @@ class TemplateInferer:
         back. Useful to avoid adding explicit hydrogens on the template which
         can prevent RDKit from finding a match between the template and the
         molecule.
+
+    .. versionadded:: 2.8.0
     """
 
     template: "Chem.Mol"
@@ -689,6 +691,8 @@ class RDKitInferer:
     """Uses RDKit's :func:`~rdkit.Chem.rdDetermineBonds.DetermineBondOrders`
     to infer bond orders and formal charges. This is the same algorithm used
     by the :ref:`xyz2mol <https://github.com/jensengroup/xyz2mol>` package.
+
+    .. versionadded:: 2.8.0
     """
 
     charge: int = 0

package/pyproject.toml

@@ -21,13 +21,13 @@ build-backend = "setuptools.build_meta"
 [project]
 name = "MDAnalysis"
 dynamic = ['version', 'readme']
-license = {file = "LICENSE"}
+license = { file = "LICENSE" }
 description = "An object-oriented toolkit to analyze molecular dynamics trajectories."
 authors = [
-    {name = 'MDAnalysis Development Team', email = 'mdanalysis@numfocus.org'}
+    { name = 'MDAnalysis Development Team', email = 'mdanalysis@numfocus.org' },
 ]
 maintainers = [
-    {name = 'MDAnalysis Core Developers', email = 'mdanalysis@numfocus.org'}
+    { name = 'MDAnalysis Core Developers', email = 'mdanalysis@numfocus.org' },
 ]
 requires-python = ">=3.10"
 dependencies = [
@@ -47,8 +47,14 @@ dependencies = [
     'mdahole2',
 ]
 keywords = [
-    "python", "science", "chemistry", "biophysics", "molecular-dynamics",
-    "computational-chemistry", "molecular-simulation", "analysis",
+    "python",
+    "science",
+    "chemistry",
+    "biophysics",
+    "molecular-dynamics",
+    "computational-chemistry",
+    "molecular-simulation",
+    "analysis",
     "trajectory-analysis",
 ]
 classifiers = [
@@ -79,7 +85,7 @@ extra_formats = [
     "pyedr>=0.7.0",
     "pytng>=0.2.3",
     "gsd>3.0.0",
-    "rdkit>=2021.09.1",
+    "rdkit>=2022.09.1",
 ]
 analysis = [
     "biopython>=1.80",
@@ -117,6 +123,4 @@ zip-safe = false
 find = {}
 
 [tool.setuptools.package-data]
-MDAnalysis = [
-    'analysis/data/*.npy',
-]
+MDAnalysis = ['analysis/data/*.npy']