Show `FutureWarning` and `DeprecationWarning` for `pytest` #4138

DanielYang59 · 2024-10-26T13:44:55Z

Summary

Show FutureWarning and DeprecationWarning for pytest, https://docs.pytest.org/en/stable/how-to/capture-warnings.html
Fix Vampire_interface (pymatgen/analysis/magnetism/heisenberg.py:323:) #4112
The warning filter tag doesn't work as expected for some reason

…rn True in future versions. Use specific 'len(elem)' or 'elem is not None' test instead.

src/pymatgen/io/vasp/outputs.py

pyproject.toml

DanielYang59 · 2024-10-27T07:55:29Z

src/pymatgen/symmetry/analyzer.py

@@ -274,7 +274,10 @@ def _get_symmetry(self) -> tuple[NDArray, NDArray]:
            "translations" gives the numpy float64 array of the translation
            vectors in scaled positions.
        """
-        dct = spglib.get_symmetry(self._cell, symprec=self._symprec, angle_tolerance=self._angle_tol)


For this we get:

tests/analysis/chemenv/connectivity/test_structure_connectivity.py::TestStructureConnectivity::test_serialization /Users/yang/developer/pymatgen/venv312/lib/python3.12/site-packages/spglib/spglib.py:640: DeprecationWarning: Use get_magnetic_symmetry() for cell with magnetic moments. warnings.warn(

But this doesn't look like a in place replace, for:

dct = spglib.get_magnetic_symmetry(self._cell, symprec=self._symprec, angle_tolerance=self._angle_tol)

I got 6 test failure:

tests/symmetry/test_analyzer.py ..F....FF.F..........F.F.......... [100%] ====================================================================== FAILURES ====================================================================== __________________________________________ TestSpacegroupAnalyzer.test_get_conventional_standard_structure ___________________________________________

src/pymatgen/alchemy/materials.py

src/pymatgen/analysis/piezo_sensitivity.py

DanielYang59 · 2024-10-27T08:09:08Z

tests/analysis/elasticity/test_elastic.py

+        # TODO: scipy.misc.central_diff_weights is deprecated in SciPy v1.10.0;
+        # and will be completely removed in SciPy v1.12.0.
+        # You may consider using findiff: https://github.com/maroba/findiff
+        # or numdifftools: https://github.com/pbrod/numdifftools
        assert_allclose(central_29, central_diff_weights(9, 2))


central_diff_weights should have been removed (but for some reason not yet), and the replacement is another package, comment this out for now?

DanielYang59 · 2024-10-27T08:17:15Z

src/pymatgen/analysis/quasiharmonic.py

-            dEdV = derivative(func, volume, dx=1e-3)
-            d2EdV2 = derivative(func, volume, dx=1e-3, n=2, order=5)
-            d3EdV3 = derivative(func, volume, dx=1e-3, n=3, order=7)
+            with warnings.catch_warnings():


Another deprecation requesting findiff or numdifftools from scipy as 6a62d63

DanielYang59 · 2024-10-27T08:46:48Z

src/pymatgen/io/lammps/inputs.py

@@ -913,14 +913,18 @@ def write_inputs(self, output_dir: str, **kwargs) -> None:
            output_dir (str): Directory to output the input files.
            **kwargs: kwargs supported by LammpsData.write_file.
        """
-        write_lammps_inputs(


Need some LAMMPS expert to fix this, this class has only one method but deprecated (the class itself is not):

pymatgen/src/pymatgen/io/lammps/inputs.py

Lines 963 to 977 in 0e65d35

class LammpsTemplateGen(TemplateInputGen):

"""Create an InputSet object for a LAMMPS run based on a template file.

The input script is constructed by substituting variables into placeholders

in the template file using python's Template.safe_substitute() function.

The data file containing coordinates and topology information can be provided

as a LammpsData instance. Alternatively, you can include a read_data command

in the template file that points to an existing data file.

Other supporting files are not handled at the moment.

To write the input files to a directory, call LammpsTemplateSet.write_input()

See pymatgen.io.template.py for additional documentation of this method.

"""

def get_input_set(

src/pymatgen/io/vasp/sets.py

tests/io/vasp/test_sets.py

DanielYang59 · 2024-10-27T09:14:10Z

tests/io/lobster/test_outputs.py

@@ -649,8 +649,8 @@ def test_attributes(self):
        atomlist = ["O1", "Mn2"]
        types = ["O", "Mn"]
        num_atoms = 2
-        assert charge_Mulliken == self.charge2.Mulliken
-        assert charge_Loewdin == self.charge2.Loewdin
+        assert charge_Mulliken == self.charge2.mulliken


@JaGeo @QuantumChemist Can you please help me have a look at this LOBTSER change 8c1cee5? A lot of test code access deprecated attribute, I wish it's not intended (to test deprecated API, expect for that one I filtered)

tests/io/lobster/test_lobsterenv.py: 206 warnings /home/runner/work/pymatgen/pymatgen/src/pymatgen/io/lobster/lobsterenv.py:193: DeprecationWarning: Mulliken is deprecated Use `mulliken` instead. self.valences = chg.Mulliken tests/io/lobster/test_lobsterenv.py: 17 warnings /home/runner/work/pymatgen/pymatgen/src/pymatgen/io/lobster/lobsterenv.py:195: DeprecationWarning: Loewdin is deprecated Use `loewdin` instead. self.valences = chg.Loewdin tests/io/lobster/test_lobsterenv.py::TestLobsterNeighbors::test_get_anion_types /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_lobsterenv.py:341: FutureWarning: get_anion_types is deprecated; use anion_types in pymatgen.io.lobster.lobsterenv instead. assert self.chem_env_lobster0_second.get_anion_types() == {Element("O")} tests/io/lobster/test_outputs.py::TestCharge::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:652: DeprecationWarning: Mulliken is deprecated Use `mulliken` instead. assert charge_Mulliken == self.charge2.Mulliken tests/io/lobster/test_outputs.py::TestCharge::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:653: DeprecationWarning: Loewdin is deprecated Use `loewdin` instead. assert charge_Loewdin == self.charge2.Loewdin tests/io/lobster/test_outputs.py: 4128 warnings /home/runner/work/pymatgen/pymatgen/src/pymatgen/io/lobster/outputs.py:1680: PendingDeprecationWarning: the matrix subclass is not the recommended way to represent matrices or deal with linear algebra (see https://docs.scipy.org/doc/numpy/user/numpy-for-matlab-users.html). Please adjust your code to use regular ndarray. self.band_overlaps_dict[spin]["matrices"].append(np.matrix(overlaps)) tests/io/lobster/test_outputs.py::TestBandoverlaps::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:1494: DeprecationWarning: bandoverlapsdict is deprecated Use `band_overlaps_dict` instead. bo_dict = self.band_overlaps1.bandoverlapsdict tests/io/lobster/test_outputs.py::TestBandoverlaps::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:1496: DeprecationWarning: bandoverlapsdict is deprecated Use `band_overlaps_dict` instead. assert self.band_overlaps1_new.bandoverlapsdict[Spin.up]["max_deviations"][10] == approx(0.0640933) tests/io/lobster/test_outputs.py::TestBandoverlaps::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:1498: DeprecationWarning: bandoverlapsdict is deprecated Use `band_overlaps_dict` instead. assert self.band_overlaps1_new.bandoverlapsdict[Spin.up]["matrices"][10].item(-1, -1) == approx(1.0) tests/io/lobster/test_outputs.py::TestBandoverlaps::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:1500: DeprecationWarning: bandoverlapsdict is deprecated Use `band_overlaps_dict` instead. assert self.band_overlaps1_new.bandoverlapsdict[Spin.up]["matrices"][10].item(0, 0) == approx(0.995849) tests/io/lobster/test_outputs.py::TestBandoverlaps::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:1503: DeprecationWarning: bandoverlapsdict is deprecated Use `band_overlaps_dict` instead. assert self.band_overlaps1_new.bandoverlapsdict[Spin.down]["max_deviations"][9] == approx(0.064369) tests/io/lobster/test_outputs.py::TestBandoverlaps::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:1505: DeprecationWarning: bandoverlapsdict is deprecated Use `band_overlaps_dict` instead. assert self.band_overlaps1_new.bandoverlapsdict[Spin.down]["matrices"][9].item(0, -1) == approx(1.37447e-09) tests/io/lobster/test_outputs.py::TestSitePotentials::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:2211: DeprecationWarning: sitepotentials_Loewdin is deprecated Use `sitepotentials_loewdin` instead. ] tests/io/lobster/test_outputs.py::TestSitePotentials::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:2218: DeprecationWarning: sitepotentials_Mulliken is deprecated Use `sitepotentials_mulliken` instead. ] tests/io/lobster/test_outputs.py::TestSitePotentials::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:2219: DeprecationWarning: madelungenergies_Loewdin is deprecated Use `madelungenergies_loewdin` instead. assert self.sitepotential.madelungenergies_Loewdin == approx(-28.64) tests/io/lobster/test_outputs.py::TestSitePotentials::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:2220: DeprecationWarning: madelungenergies_Mulliken is deprecated Use `madelungenergies_mulliken` instead. assert self.sitepotential.madelungenergies_Mulliken == approx(-40.02) tests/io/lobster/test_outputs.py::TestMadelungEnergies::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:2256: DeprecationWarning: madelungenergies_Loewdin is deprecated Use `madelungenergies_loewdin` instead. assert self.madelungenergies.madelungenergies_Loewdin == approx(-28.64) tests/io/lobster/test_outputs.py::TestMadelungEnergies::test_attributes /home/runner/work/pymatgen/pymatgen/tests/io/lobster/test_outputs.py:2257: DeprecationWarning: madelungenergies_Mulliken is deprecated Use `madelungenergies_mulliken` instead. assert self.madelungenergies.madelungenergies_Mulliken == approx(-40.02)

src/pymatgen/io/phonopy.py

DanielYang59 · 2024-10-27T11:05:27Z

src/pymatgen/analysis/magnetism/heisenberg.py

@@ -320,6 +320,8 @@ def _get_exchange_df(self):

                # Ignore the row if it is a duplicate to avoid singular matrix
                # Create a temporary DataFrame with the new row
+                ex_mat = ex_mat.dropna(how="all", axis=1)


I'm not 100% sure about this change (regarding the axis), but it should be the columns from the warning message:

/Users/yang/developer/pymatgen/src/pymatgen/analysis/magnetism/heisenberg.py:325: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation. temp_df = pd.concat([ex_mat, ex_row], ignore_index=True)

DanielYang59 · 2024-10-27T12:35:02Z

src/pymatgen/analysis/wulff.py

@@ -578,7 +578,7 @@ def get_plotly(
            x=[0],
            y=[0],
            z=[0],
-            colorbar=go.ColorBar(


I'm not experienced with plotly, could any one help me confirm this?

plotly.graph_objs.ColorBar is deprecated. Please replace it with one of the following more specific types - plotly.graph_objs.scatter.marker.ColorBar - plotly.graph_objs.surface.ColorBar - etc.

Looks like the correct colorbar for mesh3d:

pymatgen/src/pymatgen/analysis/wulff.py

Lines 575 to 589 in 0e65d35

# Add colorbar

color_scale = sorted(color_scale, key=lambda c: c[0])

colorbar = go.Mesh3d(

x=[0],

y=[0],

z=[0],

colorbar=go.ColorBar(

title={

"text": f"Surface energy {units}",

"side": "right",

"font": {"size": 25},

},

ticktext=ticktext,

tickvals=tickvals,

),

DanielYang59 · 2024-10-27T12:50:17Z

src/pymatgen/io/lobster/outputs.py

@@ -1677,7 +1677,7 @@ def _read(self, lines: list[str], spin_numbers: list[int]) -> None:
                        _lines.append(float(el))
                overlaps.append(_lines)
                if len(overlaps) == len(_lines):
-                    self.band_overlaps_dict[spin]["matrices"].append(np.matrix(overlaps))


@JaGeo would appreciate it if you could have a look at this as well, got:

/Users/yang/developer/pymatgen/src/pymatgen/io/lobster/outputs.py:1680: PendingDeprecationWarning: the matrix subclass is not the recommended way to represent matrices or deal with linear algebra (see https://docs.scipy.org/doc/numpy/user/numpy-for-matlab-users.html). Please adjust your code to use regular ndarray. self.band_overlaps_dict[spin]["matrices"].append(np.matrix(overlaps))

And as per the manual:

‘array’ or ‘matrix’? Which should I use?

Historically, NumPy has provided a special matrix type, np.matrix, which is a subclass of ndarray which makes binary operations linear algebra operations. You may see it used in some existing code instead of np.array. So, which one to use?

Short answer
Use arrays.

DanielYang59 added 6 commits October 26, 2024 21:43

show FutureWarning and DeprecationWarning

545e2a5

TODO: mark FutureWarning and DeprecationWarning for this PR

64c99d8

Merge branch 'master' into show-deprecation-warn

3a79e83

from pymatgen.core.interface import GrainBoundaryGenerator

c213125

Suppress DeprecationWarnings on pkg_resources from pybtex

bc8d0ae

suppress all warning from pybtex

52348ea

DanielYang59 force-pushed the show-deprecation-warn branch 3 times, most recently from 26ed4ec to 963dff5 Compare October 27, 2024 06:30

use single quote

0d697e9

DanielYang59 force-pushed the show-deprecation-warn branch from 963dff5 to 0d697e9 Compare October 27, 2024 06:36

DanielYang59 added 3 commits October 27, 2024 14:43

ignore Userwarning

35ad4ca

bump plotly to avoid SyntaxWarning

0aad506

DeprecationWarning: Testing an element's truth value will always retu…

0dd7e40

…rn True in future versions. Use specific 'len(elem)' or 'elem is not None' test instead.