Merge pull request #119 from lcmd-epfl/c2m-depfix

Update requirements and pyproject:
- update cell2mol version
- fix setup.py when no access to gitdir
- replace vulnerable dependencies
- add `[dev]` option
- unsure compatibility with declared numpy and pyscf versions
- adapt new functions for python 3.9

Manual checks:
- python 3.11 works
- python 3.9 works thanks to the hacks (without cell2mol)

Todo:
- drop python 3.9 after the Marvel event in January

Author: briling

pyproject.toml

@@ -37,13 +37,13 @@ classifiers = [
 # note: maximal versions are just the latest versions to exist when compatibility checks were made, later versions may also work
 requires-python = ">=3.9"
 dependencies = [
-    'numpy >= 1.22.3, < 2.4',
+    'numpy >= 1.26.1, < 2.4',
     'scipy >= 1.1.0, < 1.17',
     'tqdm >= 4.66',
 ]
 
 [project.optional-dependencies]
-pyscf = ['pyscf >= 2.0.1, < 2.12']
+pyscf = ['pyscf >= 2.1.0, < 2.12']
 qml = ['ase >= 3.22, < 3.27']
 regression = ["scikit-learn >= 0.24.2, < 1.8"]
 wigner = ["sympy >= 1.5, < 1.15"]
@@ -52,8 +52,10 @@ cell2mol = ["qstack[pyscf,cell2mol-itself]"]
 equio-itself = ["metatensor-core >= 0.1.8, == 0.1.*"]
 equio = ["qstack[pyscf,equio-itself]"]
 spahm = ["qstack[pyscf]"]
+all = ["qstack[pyscf,qml,regression,wigner,equio,gmol,cell2mol]"]
 docs-build = ["sphinx", "sphinx-argparse", "sphinx-autobuild", "myst-parser", "furo"]
-all = ["qstack[pyscf,qml,regression,wigner,equio,gmol]"]
+lint = ["ruff"]
+dev = ["qstack[all,docs-build,lint]"]
 
 [project.urls]
 Repository = "https://github.com/lcmd-epfl/Q-stack.git"

qstack/__init__.py

@@ -1,12 +1,17 @@
 """Q-stack."""
 
+import sys
+import warnings
+import builtins
 from qstack import tools
-#from qstack import compound  # needs pyscf
 from qstack import constants
 from qstack import mathutils
-#from qstack import fields  # needs scipy, pyscf
-#from qstack import basis_opt  # needs pyscf, scipy
-#from qstack import spahm  # requires pyscf
-#from qstack import orcaio  # requires pyscf
-#from qstack import qml  # requires ase
-#from qstack import regression  # requires sklearn (scikit-learn  as pip calls it)
+
+
+if sys.version_info[1]<10:
+    warnings.warn('Redefining built-in function zip for compatibility', stacklevel=1)
+    _zip = builtins.zip
+    def zip_override(*iterables, strict=False):
+        """Override built-in zip for python<3.10 to ignore `strict` argument."""
+        return _zip(*iterables)
+    builtins.zip = zip_override

qstack/fields/decomposition.py

@@ -2,7 +2,7 @@
 
 import numpy as np
 import scipy
-from pyscf import scf
+from pyscf import scf, gto
 from qstack import compound
 from . import moments
 
@@ -59,10 +59,10 @@ def get_self_repulsion(mol_or_mf, dm):
     Returns:
         float: Self-repulsion energy (a.u).
     """
-    try:
-        j, _k = mol_or_mf.get_jk()
-    except AttributeError:
+    if isinstance(mol_or_mf, gto.mole.Mole):
         j, _k = scf.hf.get_jk(mol_or_mf, dm)
+    else:
+        j, _k = mol_or_mf.get_jk()
     return np.einsum('ij,ij', j, dm)
 
 

qstack/regression/local_kernels.py

@@ -11,10 +11,10 @@
 import ctypes
 import sysconfig
 import warnings
-import itertools
 import numpy as np
 import sklearn.metrics.pairwise as _SKLEARN_PAIRWISE
 from qstack.regression import __path__ as REGMODULE_PATH
+from qstack.tools import pairwise
 
 
 RAM_BATCHING_SIZE = 1024**3 * 5  # 5GiB
@@ -66,7 +66,7 @@ def compute_distance_matrix(R1, R2):
     else:
         dists = np.zeros((batch_size, *R2.shape), dtype=dtype)
         batch_limits = np.minimum(np.arange(0, R1.shape[0]+batch_size, step=batch_size), R1.shape[0])
-        for batch_start, batch_end in itertools.pairwise(batch_limits):
+        for batch_start, batch_end in pairwise(batch_limits):
             dists_view = dists[:batch_end-batch_start]
             R1_view = R1[batch_start:batch_end, None, ...]
             np.subtract(R1_view, R2[None,:], out=dists_view)

qstack/reorder.py

@@ -51,7 +51,7 @@ def _orca2gpr_idx(l_slices, m):
         idx[s] = np.concatenate((idx[s][::-2], idx[s][1::2]))
     signs = np.ones_like(idx)
     signs[np.where(np.abs(m)>=3)] = -1  # in pyscf order
-    signs[idx] = signs  # in orca order
+    signs[idx] = np.copy(signs)  # in orca order. copy for numpy < 2
     return idx, signs
 
 

qstack/spahm/rho/utils.py

@@ -52,8 +52,8 @@ def chsp_converter(chsp):
     if fname is None:
         return np.full(n, None, dtype=object)
     if os.path.isfile(fname):
-        chsp = np.loadtxt(fname, dtype=object, converters=chsp_converter, encoding=None)
-        if len(chsp)!=n:
+        chsp = np.loadtxt(fname, dtype=object, converters={0: chsp_converter}, encoding=None)
+        if chsp.shape != (n,):
             raise RuntimeError(f'Wrong length of the file {fname}')
     else:
         raise RuntimeError(f"{fname} can not be found")

qstack/tools.py

@@ -4,11 +4,25 @@
 """
 
 import os
+import sys
 import time
 import resource
 import argparse
-import itertools
 import numpy as np
+from itertools import accumulate
+if sys.version_info[1]>=10:
+    from itertools import pairwise
+else:
+    def pairwise(iterable):
+        """Implement itertools.pairwise for python<3.10.
+
+        Taken from https://docs.python.org/3/library/itertools.html#itertools.pairwise
+        """
+        iterator = iter(iterable)
+        a = next(iterator, None)
+        for b in iterator:
+            yield a, b
+            a = b
 
 
 def unix_time_decorator(func):
@@ -115,8 +129,8 @@ def slice_generator(iterable, inc=lambda x: x, i0=0):
         tuple: (element, slice) pairs for each element in the iterable.
     """
     func = func=lambda total, elem: total+inc(elem)
-    starts = itertools.accumulate(iterable, func=func, initial=i0)
-    starts_ends = itertools.pairwise(starts)
+    starts = accumulate(iterable, func=func, initial=i0)
+    starts_ends = pairwise(starts)
     for elem, (start, end) in zip(iterable, starts_ends, strict=True):
         yield elem, np.s_[start:end]
 

requirements.py3.11.txt

@@ -1,18 +1,17 @@
 attrs==21.4.0
-certifi==2021.10.8
 h5py==3.11.0
 iniconfig==1.1.1
-packaging==21.3
+packaging==25.0
 pluggy==1.0.0
 py==1.11.0
-pyparsing==3.0.6
+pyparsing==3.2.5
 pyscf==2.5.0
 pytest==6.2.5
 numpy==2.3.*
-scipy==1.10
+scipy==1.16.3
 toml==0.10.2
-scikit-learn==1.5.0
-ase==3.22
-tqdm==4.66
-metatensor-core==0.1.8
-cell2mol @ git+https://github.com/lcmd-epfl/cell2mol.git@22473bbf12a013467137a55a63c88fbbdc95baa2
+scikit-learn==1.7.2
+ase==3.26
+tqdm==4.67
+metatensor-core==0.1.17
+cell2mol @ git+https://github.com/lcmd-epfl/cell2mol.git@40b3237

requirements.py3.13.txt

@@ -6,7 +6,7 @@ pyparsing==3.2.5
 pyscf==2.11.0
 pytest==8.4.2
 numpy==2.3.*
-scipy==1.16.2
+scipy==1.16.3
 scikit-learn==1.7.2
 ase==3.26
 tqdm==4.67

requirements.py3.9.txt

@@ -1,18 +1,16 @@
 attrs==21.4.0
-certifi==2021.10.8
 h5py==3.6.0
 iniconfig==1.1.1
 packaging==21.3
 pluggy==1.0.0
 py==1.11.0
 pyparsing==3.0.6
-pyscf==2.0.1
+pyscf>=2.1
 pytest==6.2.5
-numpy===1.22.3
+numpy>=1.26.1
 scipy==1.10
 toml==0.10.2
-scikit-learn==0.24.2
+scikit-learn==1.6.1
 ase==3.22
-tqdm==4.66
+tqdm==4.67
 metatensor-core==0.1.8
-cell2mol @ git+https://github.com/lcmd-epfl/cell2mol.git@22473bbf12a013467137a55a63c88fbbdc95baa2