Clean up test files: VASP outputs (#3653)

* move vasprun to dedicated dir

* move OUTCAR to dedicated dir

* fix vasp output file paths

* move CHGCAR to dedicated dir

* relocate DOSCAR

* relocate EIGENVAL

* relocate PROCAR

* relocate WAVECAR

* relocate XDATCAR

* relocate LOCPOT

* relocate DYNMAT

* relocate ELFCAR

* relocate CONTCAR

* relocate IBZKPT

* relocate WAVEDER and WAVEDERF

* relocate trajectory XDATCAR

* relocate OSZICAR

* temp fix: borg

* rename trajectory test files

* remove duplicate OUTCARs

* compress some CHGCAR OUTCAR PROCAR

* compress formatted WAVEDER

* pre-commit auto-fixes

* sort import

* compress OUTCAR and remove duplicate

* compress more vasp output files

* add missing gz extension

* rename some vasprun to respect extension name

* rename more vasprun to respect file extension

* remove remaining vasprun to respect file extension

* fix some paths after rename

* redirect borg to output dir for efermi search

* pre-commit auto-fixes

* fix borg error

* add test note

* fix test_xe

* rename vasprun.xe and use ScratchDir

* define VASP_OUT_DIR = f"{TEST_FILES_DIR}/vasp/outputs" for shorter file paths

---------

Signed-off-by: Haoyu (Daniel) <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Janosh Riebesell <[email protected]>

pymatgen/io/vasp/sets.py

@@ -2746,7 +2746,7 @@ def get_vasprun_outcar(path: str | Path, parse_dos: bool = True, parse_eigen: bo
     if len(vruns) == 0 or len(outcars) == 0:
         raise ValueError(f"Unable to get vasprun.xml/OUTCAR from prev calculation in {path}")
     vsfile_fullpath = str(path / "vasprun.xml")
-    outcarfile_fullpath = str(path / "OUTCAR")
+    outcarfile_fullpath = str(path / "OUTCAR.gz")
     vsfile = vsfile_fullpath if vsfile_fullpath in vruns else sorted(vruns)[-1]
     outcarfile = outcarfile_fullpath if outcarfile_fullpath in outcars else sorted(outcars)[-1]
     return (

pymatgen/util/testing/__init__.py

@@ -26,6 +26,7 @@
 MODULE_DIR = Path(__file__).absolute().parent
 STRUCTURES_DIR = MODULE_DIR / ".." / "structures"
 TEST_FILES_DIR = Path(SETTINGS.get("PMG_TEST_FILES_DIR", f"{ROOT}/tests/files"))
+VASP_OUT_DIR = f"{TEST_FILES_DIR}/vasp/outputs"
 # fake POTCARs have original header information, meaning properties like number of electrons,
 # nuclear charge, core radii, etc. are unchanged (important for testing) while values of the and
 # pseudopotential kinetic energy corrections are scrambled to avoid VASP copyright infringement

tests/analysis/test_structure_analyzer.py

@@ -18,12 +18,12 @@
 )
 from pymatgen.core import Element, Lattice, Structure
 from pymatgen.io.vasp.outputs import Xdatcar
-from pymatgen.util.testing import TEST_FILES_DIR, PymatgenTest
+from pymatgen.util.testing import TEST_FILES_DIR, VASP_OUT_DIR, PymatgenTest
 
 
 class TestVoronoiAnalyzer(PymatgenTest):
     def setUp(self):
-        self.ss = Xdatcar(f"{TEST_FILES_DIR}/XDATCAR.MD").structures
+        self.ss = Xdatcar(f"{VASP_OUT_DIR}/XDATCAR.MD").structures
         self.s = self.ss[1]
         self.va = VoronoiAnalyzer(cutoff=4)
 
@@ -40,7 +40,7 @@ def test_analyze(self):
 class TestRelaxationAnalyzer(unittest.TestCase):
     def setUp(self):
         s1 = Structure.from_file(f"{TEST_FILES_DIR}/POSCAR.Li2O")
-        s2 = Structure.from_file(f"{TEST_FILES_DIR}/CONTCAR.Li2O")
+        s2 = Structure.from_file(f"{VASP_OUT_DIR}/CONTCAR.Li2O")
         self.analyzer = RelaxationAnalyzer(s1, s2)
 
     def test_vol_and_para_changes(self):
@@ -77,7 +77,7 @@ def test_connectivity_array(self):
 
 class TestMiscFunction(PymatgenTest):
     def test_average_coordination_number(self):
-        xdatcar = Xdatcar(f"{TEST_FILES_DIR}/XDATCAR.MD")
+        xdatcar = Xdatcar(f"{VASP_OUT_DIR}/XDATCAR.MD")
         coordination_numbers = average_coordination_number(xdatcar.structures, freq=1)
         assert coordination_numbers["Fe"] == approx(
             4.771903318390836, 5

tests/analysis/test_xps.py

@@ -2,12 +2,12 @@
 
 from pymatgen.analysis.xps import XPS
 from pymatgen.io.vasp import Vasprun
-from pymatgen.util.testing import TEST_FILES_DIR, PymatgenTest
+from pymatgen.util.testing import VASP_OUT_DIR, PymatgenTest
 
 
 class XPSTestCase(PymatgenTest):
     def test_from_dos(self):
-        vasp_run = Vasprun(f"{TEST_FILES_DIR}/vasprun.xml.LiF.gz")
+        vasp_run = Vasprun(f"{VASP_OUT_DIR}/vasprun.LiF.xml.gz")
         dos = vasp_run.complete_dos
         xps = XPS.from_dos(dos)
         assert len(xps) == 301

tests/apps/borg/test_hive.py

@@ -3,6 +3,7 @@
 import os
 import unittest
 
+from monty.tempfile import ScratchDir
 from pytest import approx
 
 from pymatgen.apps.borg.hive import (
@@ -11,7 +12,7 @@
     VaspToComputedEntryDrone,
 )
 from pymatgen.entries.computed_entries import ComputedStructureEntry
-from pymatgen.util.testing import TEST_FILES_DIR
+from pymatgen.util.testing import TEST_FILES_DIR, VASP_OUT_DIR
 
 
 class TestVaspToComputedEntryDrone(unittest.TestCase):
@@ -20,23 +21,30 @@ def setUp(self):
         self.structure_drone = VaspToComputedEntryDrone(inc_structure=True)
 
     def test_get_valid_paths(self):
-        for path in os.walk(TEST_FILES_DIR):
-            if path[0] == TEST_FILES_DIR:
+        for path in os.walk(VASP_OUT_DIR):
+            if path[0] == VASP_OUT_DIR:
                 assert len(self.drone.get_valid_paths(path)) > 0
 
     def test_assimilate(self):
-        entry = self.drone.assimilate(TEST_FILES_DIR)
-        for param in ("hubbards", "is_hubbard", "potcar_spec", "run_type"):
-            assert param in entry.parameters
-        assert entry.data["efermi"] == approx(-6.62148548)
-        assert entry.reduced_formula == "Xe"
-        assert entry.energy == approx(0.5559329)
-        entry = self.structure_drone.assimilate(TEST_FILES_DIR)
-        assert entry.reduced_formula == "Xe"
-        assert entry.energy == approx(0.5559329)
-        assert isinstance(entry, ComputedStructureEntry)
-        assert entry.structure is not None
-        # assert len(entry.parameters["history"]) == 2
+        """Test assimilate data from "vasprun.xe.xml.gz" file."""
+
+        with ScratchDir("."):
+            # Need to rename the test file to "vasprun.xml.xe.gz" as
+            # hive is looking for pattern "vasprun.xml*"
+            os.symlink(f"{VASP_OUT_DIR}/vasprun.xe.xml.gz", "vasprun.xml.xe.gz")
+            entry = self.drone.assimilate(".")
+
+            for param in ("hubbards", "is_hubbard", "potcar_spec", "run_type"):
+                assert param in entry.parameters
+            assert entry.data["efermi"] == approx(-6.62148548)
+            assert entry.reduced_formula == "Xe"
+            assert entry.energy == approx(0.5559329)
+
+            entry = self.structure_drone.assimilate(".")
+            assert entry.reduced_formula == "Xe"
+            assert entry.energy == approx(0.5559329)
+            assert isinstance(entry, ComputedStructureEntry)
+            assert entry.structure is not None
 
     def test_as_from_dict(self):
         dct = self.structure_drone.as_dict()
@@ -50,8 +58,8 @@ def setUp(self):
         self.structure_drone = SimpleVaspToComputedEntryDrone(inc_structure=True)
 
     def test_get_valid_paths(self):
-        for path in os.walk(TEST_FILES_DIR):
-            if path[0] == TEST_FILES_DIR:
+        for path in os.walk(VASP_OUT_DIR):
+            if path[0] == VASP_OUT_DIR:
                 assert len(self.drone.get_valid_paths(path)) > 0
 
     def test_as_from_dict(self):

tests/command_line/test_bader_caller.py

@@ -12,7 +12,7 @@
 from pytest import approx
 
 from pymatgen.command_line.bader_caller import BaderAnalysis, bader_analysis_from_path
-from pymatgen.util.testing import TEST_FILES_DIR, PymatgenTest
+from pymatgen.util.testing import TEST_FILES_DIR, VASP_OUT_DIR, PymatgenTest
 
 
 @unittest.skipIf(not which("bader"), "bader executable not present")
@@ -23,9 +23,9 @@ def setUp(self):
     def test_init(self):
         # test with reference file
         analysis = BaderAnalysis(
-            chgcar_filename=f"{TEST_FILES_DIR}/CHGCAR.Fe3O4",
+            chgcar_filename=f"{VASP_OUT_DIR}/CHGCAR.Fe3O4.gz",
             potcar_filename=f"{TEST_FILES_DIR}/POTCAR.Fe3O4",
-            chgref_filename=f"{TEST_FILES_DIR}/CHGCAR.Fe3O4_ref",
+            chgref_filename=f"{VASP_OUT_DIR}/CHGCAR.Fe3O4_ref.gz",
         )
         assert len(analysis.data) == 14
         assert analysis.data[0]["charge"] == approx(6.6136782, abs=1e-3)
@@ -55,7 +55,7 @@ def test_init(self):
         assert struct[0].specie.oxi_state == approx(1.3863218, abs=1e-3)
 
         # make sure bader still runs without reference file
-        analysis = BaderAnalysis(chgcar_filename=f"{TEST_FILES_DIR}/CHGCAR.Fe3O4")
+        analysis = BaderAnalysis(chgcar_filename=f"{VASP_OUT_DIR}/CHGCAR.Fe3O4.gz")
         assert len(analysis.data) == 14
 
         # Test Cube file format parsing
@@ -118,9 +118,9 @@ def test_bader_analysis_from_path(self):
     def test_atom_parsing(self):
         # test with reference file
         analysis = BaderAnalysis(
-            chgcar_filename=f"{TEST_FILES_DIR}/CHGCAR.Fe3O4",
+            chgcar_filename=f"{VASP_OUT_DIR}/CHGCAR.Fe3O4.gz",
             potcar_filename=f"{TEST_FILES_DIR}/POTCAR.Fe3O4",
-            chgref_filename=f"{TEST_FILES_DIR}/CHGCAR.Fe3O4_ref",
+            chgref_filename=f"{VASP_OUT_DIR}/CHGCAR.Fe3O4_ref.gz",
             parse_atomic_densities=True,
         )
 
@@ -139,4 +139,4 @@ def test_missing_file_bader_exe_path(self):
                 RuntimeError, match="BaderAnalysis requires the executable bader be in the PATH or the full path "
             ),
         ):
-            BaderAnalysis(chgcar_filename=f"{TEST_FILES_DIR}/CHGCAR.Fe3O4", bader_path="")
+            BaderAnalysis(chgcar_filename=f"{VASP_OUT_DIR}/CHGCAR.Fe3O4.gz", bader_exe_path="")

tests/core/test_trajectory.py

@@ -10,13 +10,13 @@
 from pymatgen.core.trajectory import Trajectory
 from pymatgen.io.qchem.outputs import QCOutput
 from pymatgen.io.vasp.outputs import Xdatcar
-from pymatgen.util.testing import TEST_FILES_DIR, PymatgenTest
+from pymatgen.util.testing import TEST_FILES_DIR, VASP_OUT_DIR, PymatgenTest
 
 
 class TestTrajectory(PymatgenTest):
     def setUp(self):
-        xdatcar = Xdatcar(f"{TEST_FILES_DIR}/Traj_XDATCAR")
-        self.traj = Trajectory.from_file(f"{TEST_FILES_DIR}/Traj_XDATCAR")
+        xdatcar = Xdatcar(f"{VASP_OUT_DIR}/XDATCAR_traj")
+        self.traj = Trajectory.from_file(f"{VASP_OUT_DIR}/XDATCAR_traj")
         self.structures = xdatcar.structures
 
         out = QCOutput(f"{TEST_FILES_DIR}/molecules/new_qchem_files/ts.out")
@@ -221,15 +221,15 @@ def test_extend(self):
         traj = copy.deepcopy(self.traj)
 
         # Case of compatible trajectories
-        compatible_traj = Trajectory.from_file(f"{TEST_FILES_DIR}/Traj_Combine_Test_XDATCAR_1")
+        compatible_traj = Trajectory.from_file(f"{VASP_OUT_DIR}/XDATCAR_traj_combine_test_1")
         traj.extend(compatible_traj)
 
-        full_traj = Trajectory.from_file(f"{TEST_FILES_DIR}/Traj_Combine_Test_XDATCAR_Full")
+        full_traj = Trajectory.from_file(f"{VASP_OUT_DIR}/XDATCAR_traj_combine_test_full")
         compatible_success = self._check_traj_equality(self.traj, full_traj)
 
         # Case of incompatible trajectories
         traj = copy.deepcopy(self.traj)
-        incompatible_traj = Trajectory.from_file(f"{TEST_FILES_DIR}/Traj_Combine_Test_XDATCAR_2")
+        incompatible_traj = Trajectory.from_file(f"{VASP_OUT_DIR}/XDATCAR_traj_combine_test_2")
         incompatible_test_success = False
         try:
             traj.extend(incompatible_traj)

tests/electronic_structure/test_bandstructure.py

@@ -20,7 +20,7 @@
 from pymatgen.electronic_structure.core import Orbital, Spin
 from pymatgen.electronic_structure.plotter import BSPlotterProjected
 from pymatgen.io.vasp import BSVasprun
-from pymatgen.util.testing import TEST_FILES_DIR, PymatgenTest
+from pymatgen.util.testing import TEST_FILES_DIR, VASP_OUT_DIR, PymatgenTest
 
 
 class TestKpoint(unittest.TestCase):
@@ -264,7 +264,7 @@ def test_reconstruct_band_structure(self):
 
     def test_vasprun_bs(self):
         bsv = BSVasprun(
-            f"{TEST_FILES_DIR}/vasprun.xml.gz",
+            f"{VASP_OUT_DIR}/vasprun.xml.gz",
             parse_projected_eigen=True,
             parse_potcar_file=True,
         )

tests/electronic_structure/test_plotter.py

@@ -30,7 +30,7 @@
     plot_ellipsoid,
 )
 from pymatgen.io.vasp import Vasprun
-from pymatgen.util.testing import TEST_FILES_DIR, PymatgenTest
+from pymatgen.util.testing import TEST_FILES_DIR, VASP_OUT_DIR, PymatgenTest
 
 rc("text", usetex=False)  # Disabling latex is needed for this test to work.
 
@@ -212,7 +212,7 @@ class TestBSDOSPlotter(unittest.TestCase):
     # Minimal baseline testing for get_plot. not a true test. Just checks that
     # it can actually execute.
     def test_methods(self):
-        vasp_run = Vasprun(f"{TEST_FILES_DIR}/vasprun_Si_bands.xml.gz")
+        vasp_run = Vasprun(f"{VASP_OUT_DIR}/vasprun_Si_bands.xml.gz")
         plotter = BSDOSPlotter()
         band_struct = vasp_run.get_band_structure(kpoints_filename=f"{TEST_FILES_DIR}/KPOINTS_Si_bands")
         ax = plotter.get_plot(band_struct)

tests/entries/test_computed_entries.py

@@ -22,10 +22,9 @@
     TemperatureEnergyAdjustment,
 )
 from pymatgen.io.vasp.outputs import Vasprun
-from pymatgen.util.testing import TEST_FILES_DIR
+from pymatgen.util.testing import TEST_FILES_DIR, VASP_OUT_DIR
 
-filepath = f"{TEST_FILES_DIR}/vasprun.xml.gz"
-vasp_run = Vasprun(filepath)
+vasp_run = Vasprun(f"{VASP_OUT_DIR}/vasprun.xml.gz")
 
 
 def test_energy_adjustment():