ci: add pyright checker (#215)

Signed-off-by: Jinzhe Zeng <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

.github/workflows/pyright.yml

@@ -0,0 +1,27 @@
+on:
+  push:
+    branches:
+      - master
+    tags:
+      - v*
+  pull_request:
+  merge_group:
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref || github.run_id }}
+  cancel-in-progress: true
+
+name: Type checker
+jobs:
+  test:
+    name: pyright
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+    - uses: actions/setup-python@v5
+      with:
+        python-version: '3.12'
+    - run: pipx run uv pip install --system -e .
+    - uses: jakebailey/pyright-action@v2
+      with:
+        version: 1.1.364

mddatasetbuilder/datasetbuilder.py

@@ -21,15 +21,17 @@
 import tempfile
 import time
 from collections import Counter, defaultdict
+from typing import List, Optional
 
 import numpy as np
+from ase.atoms import Atoms
 from ase.data import atomic_numbers
 from ase.io import write as write_xyz
 from sklearn import preprocessing
 from sklearn.cluster import MiniBatchKMeans
 
 from ._logger import logger
-from .detect import Detect
+from .detect import Detect, DetectDump
 from .utils import (
     bytestolist,
     listtobytes,
@@ -97,7 +99,7 @@ def __init__(
         nproc=None,
         pbc=True,
         fragment=False,
-        errorfilename=None,
+        errorfilename: Optional[List[str]] = None,
         errorlimit=0.0,
         atom_pref=False,
     ):
@@ -302,13 +304,15 @@ def _writestepmatrix(self, item):
         step, lines = item
         results = []
         if step in self.dstep:
+            assert isinstance(self.crddetector, DetectDump)
             step_atoms, _ = self.crddetector.readcrd(lines)
             for atoma in self.dstep[step]:
                 # atom ID starts from 1
                 distances = step_atoms.get_distances(
                     atoma - 1, range(len(step_atoms)), mic=True
                 )
                 cutoffatoms = step_atoms[distances < self.cutoff]
+                assert isinstance(cutoffatoms, Atoms)
                 symbols = cutoffatoms.get_chemical_symbols()
                 results.append(
                     (
@@ -364,7 +368,9 @@ def _clusterdatas(cls, X, n_clusters, n_each=1):
         min_max_scaler = preprocessing.MinMaxScaler()
         X = np.array(min_max_scaler.fit_transform(X))
         clus = MiniBatchKMeans(
-            n_clusters=n_clusters, init_size=(min(3 * n_clusters, len(X))), n_init=3
+            n_clusters=n_clusters,
+            init_size=(min(3 * n_clusters, len(X))),
+            n_init=3,  # type: ignore
         )
         labels = clus.fit_predict(X)
         choosedidx = []
@@ -477,6 +483,7 @@ def _convertgjf(self, gjffilename, takenatomidindex, atoms_whole):
         multiplicity_whole = sum(multiplicities) - len(takenatomidindex) + 1
         multiplicity_whole_str = f"0 {multiplicity_whole}"
         title = "\nGenerated by MDDatasetMaker (Author: Jinzhe Zeng)\n"
+        connect = None
         if len(self.qmkeywords) > 1:
             connect = "\n--link1--\n"
             chk = [f"%chk={os.path.splitext(os.path.basename(gjffilename))[0]}.chk"]
@@ -505,6 +512,7 @@ def _convertgjf(self, gjffilename, takenatomidindex, atoms_whole):
                     ]
                 )
         for kw in itertools.islice(self.qmkeywords, 1, None):
+            assert connect is not None
             buff.extend((connect, *chk, kw, title, f"0 {multiplicity_whole}", "\n"))
         buff.append("\n")
         with open(gjffilename, "w") as f:
@@ -530,10 +538,12 @@ def _writestepxyzfile(self, item):
         results = 0
         if step in self.dstep:
             if len(lines) == 2:
+                assert isinstance(self.crddetector, DetectDump)
                 step_atoms, _ = self.crddetector.readcrd(lines[0])
-                molecules = self.bonddetector.readmolecule(lines[1])
+                molecules, _ = self.bonddetector.readmolecule(lines[1])
             else:
                 molecules, step_atoms = self.bonddetector.readmolecule(lines)
+            assert step_atoms is not None
             for atoma, trajatomfilename, itype, itotal in self.dstep[step]:
                 # update counter
                 folder = str(itotal // 1000).zfill(self.foldermaxlength)
@@ -555,9 +565,10 @@ def _writestepxyzfile(self, item):
                         idsum += len(mol_atomid)
                 idx = np.concatenate(takenatomids)
                 cutoffatoms = step_atoms[idx]
-                cutoffatoms[np.nonzero(idx == atoma - 1)[0][0]].tag = 1
+                assert isinstance(cutoffatoms, Atoms)
+                cutoffatoms[np.nonzero(idx == atoma - 1)[0][0]].tag = 1  # type: ignore
                 cutoffatoms.wrap(
-                    center=step_atoms[atoma - 1].position
+                    center=step_atoms[atoma - 1].position  # type: ignore
                     / cutoffatoms.get_cell_lengths_and_angles()[0:3],
                     pbc=cutoffatoms.get_pbc(),
                 )
@@ -632,6 +643,7 @@ def erroriter(self):
         str
             the line of model deviation
         """
+        assert self.errorfilename is not None
         fns = must_be_list(self.errorfilename)
         for fn in fns:
             with open(fn) as f:

mddatasetbuilder/deepmd.py

@@ -53,7 +53,7 @@ def _searchpath(self):
                 self._preparedeepmdforLOG, tqdm(logfiles, disable=None)
             ):
                 multi_systems.append(system)
-        multi_systems.to_deepmd_npy(self.deepmd_dir)
+        multi_systems.to_deepmd_npy(self.deepmd_dir)  # type: ignore
         for formula, system in multi_systems.systems.items():
             self.system_paths.append(os.path.join(self.deepmd_dir, formula))
             self.batch_size.append(

mddatasetbuilder/detect.py

@@ -4,12 +4,13 @@
 from abc import ABCMeta, abstractmethod
 from collections import defaultdict
 from enum import Enum, auto
+from typing import List, Optional, Tuple, Union, cast
 
 import numpy as np
 from ase import Atom, Atoms
 from openbabel import openbabel
 
-from .dps import dps as connectmolecule
+from mddatasetbuilder.dps import dps as connectmolecule
 
 
 class Detect(metaclass=ABCMeta):
@@ -24,16 +25,16 @@ def __init__(self, filename, atomname, pbc, errorlimit=None, errorfilename=None)
         self.steplinenum = self._readN()
 
     @abstractmethod
-    def _readN(self):
+    def _readN(self) -> int:
         pass
 
     @abstractmethod
-    def readatombondtype(self, item):
+    def readatombondtype(self, item) -> Tuple[dict, int]:
         """Read bond types of atoms such as C1111."""
         pass
 
     @abstractmethod
-    def readmolecule(self, lines):
+    def readmolecule(self, lines) -> Tuple[List[List[int]], Optional[Atoms]]:
         """Read molecules."""
         pass
 
@@ -54,6 +55,10 @@ class DetectBond(Detect):
 
     def _readN(self):
         """Read bondfile N, which should be at very beginning."""
+        N = None
+        atomtype = None
+        stepaindex = None
+        stepbindex = None
         # copy from reacnetgenerator on 2018-12-15
         with open(
             self.filename if isinstance(self.filename, str) else self.filename[0]
@@ -72,7 +77,10 @@ def _readN(self):
                         atomtype = np.zeros(N, dtype=int)
                 else:
                     s = line.split()
+                    assert atomtype is not None
                     atomtype[int(s[0]) - 1] = int(s[1])
+        if stepaindex is None or stepbindex is None or N is None or atomtype is None:
+            raise RuntimeError("The bond file is not completed")
         steplinenum = stepbindex - stepaindex
         self._N = N
         self.atomtype = atomtype
@@ -110,7 +118,7 @@ def readatombondtype(self, item):
                     )
         return d, step
 
-    def readmolecule(self, lines):
+    def readmolecule(self, lines) -> Tuple[List[List[int]], Optional[Atoms]]:
         """Return molecules from lines.
 
         Parameters
@@ -122,16 +130,19 @@ def readmolecule(self, lines):
         -------
         molecules: list
             Indexes of atoms in molecules.
+        None
+            None
         """
         # copy from reacnetgenerator on 2018-12-15
-        bond = [None] * self._N
+        bond: List[Optional[List[int]]] = [None] * self._N
         for line in lines:
             if line:
                 if not line.startswith("#"):
                     s = line.split()
                     bond[int(s[0]) - 1] = [int(x) - 1 for x in s[3 : 3 + int(s[2])]]
-        molecules = connectmolecule(bond)
-        return molecules
+        bond_ = cast(List[List[int]], bond)
+        molecules = connectmolecule(bond_)
+        return molecules, None
 
 
 class DetectDump(Detect):
@@ -140,9 +151,14 @@ class DetectDump(Detect):
     def _readN(self):
         # copy from reacnetgenerator on 2018-12-15
         iscompleted = False
+        N = None
+        atomtype = None
+        stepaindex = None
+        stepbindex = None
         with open(
             self.filename if isinstance(self.filename, str) else self.filename[0]
         ) as f:
+            linecontent = None
             for index, line in enumerate(f):
                 if line.startswith("ITEM:"):
                     linecontent = self.LineType.linecontent(line)
@@ -154,7 +170,9 @@ def _readN(self):
                         self.yidx = keys.index("y") - 2
                         self.zidx = keys.index("z") - 2
                 else:
-                    if linecontent == self.LineType.NUMBER:
+                    if linecontent is None:
+                        raise RuntimeError("No ITEM: in the dump file")
+                    elif linecontent == self.LineType.NUMBER:
                         if iscompleted:
                             stepbindex = index
                             break
@@ -165,7 +183,10 @@ def _readN(self):
                         atomtype = np.zeros(N, dtype=int)
                     elif linecontent == self.LineType.ATOMS:
                         s = line.split()
+                        assert atomtype is not None
                         atomtype[int(s[self.id_idx]) - 1] = int(s[self.tidx])
+        if stepaindex is None or stepbindex is None or N is None or atomtype is None:
+            raise RuntimeError("The dump file is not completed")
         steplinenum = stepbindex - stepaindex
         self._N = N
         self.atomtype = atomtype
@@ -188,21 +209,23 @@ def readatombondtype(self, item):
             the step index
         """
         (step, lines), needlerror = item
-        if needlerror:
-            trajline, errorline = lines
-            lerror = np.fromstring(errorline, dtype=float, sep=" ")[7:]
+        lerror: Optional[Union[np.ndarray, List[float]]] = None
         d = defaultdict(list)
         step_atoms, ids = self.readcrd(lines)
         if needlerror:
+            trajline, errorline = lines
+            lerror = np.fromstring(errorline, dtype=float, sep=" ")[7:]
             lerror = [x for (y, x) in sorted(zip(ids, lerror))]
         level = self._crd2bond(step_atoms, readlevel=True)
         for i, (n, l) in enumerate(zip(self.atomnames, level)):
-            if not needlerror or lerror[i] > self.errorlimit:
+            if lerror is None or (
+                self.errorlimit is not None and lerror[i] > self.errorlimit
+            ):
                 # Note that atom id starts from 1
                 d[pickle.dumps((n, sorted(l)))].append(i + 1)
         return d, step
 
-    def readmolecule(self, lines):
+    def readmolecule(self, lines) -> Tuple[List[List[int]], Optional[Atoms]]:
         """Return molecules from lines.
 
         Parameters
@@ -250,11 +273,11 @@ def _crd2bond(cls, step_atoms, readlevel):
             mol.CloneData(uc)
             mol.SetPeriodicMol()
         mol.ConnectTheDots()
-        if not readlevel:
-            bond = [[] for i in range(atomnumber)]
-        else:
+        # when readlevel is False, bond is used to store connected atoms
+        # otherwise, bondlevel is used to store bond orders
+        bond = [[] for i in range(atomnumber)]
+        if readlevel:
             mol.PerceiveBondOrders()
-            bondlevel = [[] for i in range(atomnumber)]
         mol.EndModify()
         for b in openbabel.OBMolBondIter(mol):
             s1 = b.GetBeginAtom().GetId()
@@ -264,23 +287,26 @@ def _crd2bond(cls, step_atoms, readlevel):
                 bond[s2].append(s1)
             else:
                 level = b.GetBondOrder()
-                bondlevel[s1].append(level)
-                bondlevel[s2].append(level)
-        return bondlevel if readlevel else bond
+                bond[s1].append(level)
+                bond[s2].append(level)
+        return bond
 
-    def readcrd(self, item):
+    def readcrd(self, item) -> Tuple[Atoms, List[int]]:
         """Only this function can read coordinates."""
         lines = item
         # box information
         ss = []
         step_atoms = []
         ids = []
+        linecontent = None
         for line in lines:
             if line:
                 if line.startswith("ITEM:"):
                     linecontent = self.LineType.linecontent(line)
                 else:
-                    if linecontent == self.LineType.ATOMS:
+                    if linecontent is None:
+                        raise RuntimeError("No ITEM: in the dump file")
+                    elif linecontent == self.LineType.ATOMS:
                         s = line.split()
                         ids.append(int(s[self.id_idx]))
                         step_atoms.append(
@@ -314,8 +340,8 @@ def readcrd(self, item):
             [[xhi - xlo, 0.0, 0.0], [xy, yhi - ylo, 0.0], [xz, yz, zhi - zlo]]
         )
         # sort by ID
-        step_atoms = [x for (y, x) in sorted(zip(ids, step_atoms))]
-        step_atoms = Atoms(step_atoms, cell=boxsize, pbc=self.pbc)
+        step_atoms_ = [x for (y, x) in sorted(zip(ids, step_atoms))]
+        step_atoms = Atoms(step_atoms_, cell=boxsize, pbc=self.pbc)
         return step_atoms, ids
 
     class LineType(Enum):

mddatasetbuilder/dps.pyi

@@ -0,0 +1 @@
+def dps(bonds: list[list[int]]) -> list[list[int]]: ...

mddatasetbuilder/utils.py

@@ -3,7 +3,7 @@
 import itertools
 import pickle
 from multiprocessing import Pool, Semaphore
-from typing import BinaryIO, Union
+from typing import BinaryIO, List, TypeVar, Union, overload
 
 import lz4.frame
 from tqdm.auto import tqdm
@@ -192,7 +192,7 @@ def bytestolist(x):
     object
         The decompressed object.
     """
-    return pickle.loads(decompress(x, isbytes=True))
+    return pickle.loads(decompress(x, isbytes=True))  # type: ignore
 
 
 def run_mp(nproc, **arg):
@@ -231,7 +231,16 @@ def run_mp(nproc, **arg):
         pool.join()
 
 
-def must_be_list(obj):
+T = TypeVar("T")
+
+
+@overload
+def must_be_list(obj: List[T]) -> List[T]: ...
+@overload
+def must_be_list(obj: T) -> List[T]: ...
+
+
+def must_be_list(obj: Union[T, List[T]]) -> List[T]:
     """Convert a object to a list if the object is not a list.
 
     Parameters