Make tests pytest parametrized

docs/source/conf.py

@@ -38,7 +38,7 @@
 numpydoc_validation_exclude = {
     r"\.__weakref__$",
     r"\.__repr__$",
-    r"janus_core\.janus_types",
+    r"janus_core\.helpers\.janus_types",
 }
 numpydoc_class_members_toctree = False
 
@@ -192,4 +192,6 @@
 nitpick_ignore = [
     ("py:class", "Logger"),
     ("py:class", "numpy.float64"),
+    ("py:class", "ellipsis"),
+    ("py:class", "T")
 ]

janus_core/helpers/stats.py

@@ -4,13 +4,20 @@
 
 from functools import singledispatchmethod
 import re
-from types import EllipsisType
+from typing import Iterator, TypeVar
 
 from numpy import float64, genfromtxt, zeros
 from numpy.typing import NDArray
 
 from janus_core.helpers.janus_types import PathLike
 
+try:  # Python >=3.10
+    from types import EllipsisType
+except ImportError:
+    EllipsisType = type(...)
+
+T = TypeVar("T")
+
 
 class Stats:
     """
@@ -41,29 +48,82 @@ def __init__(self, source: PathLike) -> None:
         self.read()
 
     @singledispatchmethod
-    def __getitem__(self, ind):
+    def _getind(self, lab: T) -> T:  # pylint: disable=no-self-use
+        """
+        Convert an index label from str to int if present in labels.
+
+        Otherwise return the input.
+
+        Parameters
+        ----------
+        lab : str
+            Label to find.
+
+        Returns
+        -------
+        int
+            Index of label in self or input if not string.
+
+        Raises
+        ------
+        IndexError
+            Label not found in labels.
+        """
+        return lab
+
+    @_getind.register
+    def _(self, lab: str) -> int:
+        # Case-insensitive fuzzy match, only has to be `in` the labels
+        index = next(
+            (
+                index
+                for index, label in enumerate(self.labels)
+                if lab.lower() in label.lower()
+            ),
+            None,
+        )
+        if index is None:
+            raise IndexError(f"{lab} not found in labels")
+        return index
+
+    @singledispatchmethod
+    def __getitem__(self, ind) -> NDArray[float64]:
+        """
+        Get member of stats data by label or index.
+
+        Parameters
+        ----------
+        ind : Any
+            Index or label to find.
+
+        Returns
+        -------
+        NDArray[float64]
+            Columns of data by label.
+
+        Raises
+        ------
+        IndexError
+            Invalid index type or label not found in labels.
+        """
         raise IndexError(f"Unknown index {ind}")
 
     @__getitem__.register(int)
     @__getitem__.register(slice)
     @__getitem__.register(EllipsisType)
-    def _(self, ind):
+    def _(self, ind) -> NDArray[float64]:
         return self.data[:, ind]
 
+    @__getitem__.register(list)
     @__getitem__.register(tuple)
-    def _(self, ind):
-        return self.data[ind]
+    def _(self, ind) -> NDArray[float64]:
+        ind = list(map(self._getind, ind))
+        return self.data[:, ind]
 
     @__getitem__.register(str)
-    def _(self, ind):
-        # Case-insensitive fuzzy match, only has to be `in` the labels
-        index = next((index
-                      for index, label in enumerate(self.labels)
-                      if ind.lower() in label.lower()),
-                     None)
-        if index is None:
-            raise IndexError(f"{ind} not found in labels")
-        return self[index]
+    def _(self, ind) -> NDArray[float64]:  # numpydoc ignore=GL08
+        ind = self._getind(ind)
+        return self[ind]
 
     @property
     def rows(self) -> int:
@@ -137,6 +197,18 @@ def data(self) -> NDArray[float64]:
         """
         return self._data
 
+    @property
+    def data_tags(self) -> Iterator[tuple[str, str]]:
+        """
+        Return the labels and their units together.
+
+        Returns
+        -------
+        Iterator[tuple[str, str]]
+            Zipped labels and units
+        """
+        return zip(self.labels, self.units)
+
     def read(self) -> None:
         """
         Read MD stats and store them in `data`.
@@ -161,6 +233,6 @@ def __repr__(self) -> str:
 
         header = f"contains {self.columns} timeseries, each with {self.rows} elements"
         header += "\nindex label units"
-        for index, (label, unit) in enumerate(zip(self.labels, self.units)):
+        for index, (label, unit) in enumerate(self.data_tags):
             header += f"\n{index} {label} {unit}"
         return header

tests/test_stats.py

@@ -2,37 +2,81 @@
 
 from pathlib import Path
 
+import pytest
 from pytest import approx
 
 from janus_core.helpers.stats import Stats
 
 DATA_PATH = Path(__file__).parent / "data"
 
 
-def test_stats(capsys):
-    """Test reading md stats."""
-    data_path = DATA_PATH / "md-stats.dat"
-
-    stat_data = Stats(data_path)
-
-    assert stat_data.rows == 100
-    assert stat_data.columns == 18
-    assert stat_data.data[99, 17] == approx(300.0)
-    assert stat_data.units[0] == ""
-    assert stat_data.units[17] == "K"
-    assert stat_data.labels[0] == "# Step"
-    assert stat_data.labels[17] == "Target T"
-
-    # Check getitem form
-    assert (stat_data["target t"] == stat_data.data[:, 17]).all()
-    assert (stat_data[:, 17] == stat_data.data[:, 17]).all()
-    assert (stat_data[17] == stat_data.data[:, 17]).all()
-
-    print(stat_data)
-    std_out_err = capsys.readouterr()
-    assert std_out_err.err == ""
-    assert "index label units" in std_out_err.out
-    assert (
-        f"contains {stat_data.columns} timeseries, each with {stat_data.rows} elements"
-        in std_out_err.out
+class TestStats:
+    """Tests for the stats type"""
+
+    data = Stats(DATA_PATH / "md-stats.dat")
+
+    @pytest.mark.parametrize(
+        "attr,expected",
+        (
+            ("rows", 100),
+            ("columns", 18),
+        ),
+        ids=(
+            "get rows",
+            "get_cols",
+        ),
+    )
+    def test_props(self, attr, expected):
+        assert getattr(self.data, attr) == expected
+
+    @pytest.mark.parametrize(
+        "attr,ind,expected",
+        (
+            ("data", (99, 17), approx(300.0)),
+            ("units", 0, ""),
+            ("units", 17, "K"),
+            ("labels", 0, "# Step"),
+            ("labels", 17, "Target T"),
+        ),
+        ids=(
+            "data value",
+            "Step units",
+            "Target T units",
+            "Step label",
+            "Target T label",
+        ),
     )
+    def test_data_index(self, attr, ind, expected):
+        assert getattr(self.data, attr)[ind] == expected
+
+    @pytest.mark.parametrize(
+        "ind,expectedcol",
+        (
+            ("target t", 17),
+            (17, 17),
+            (slice(3, 7, 2), (3, 5)),
+            ((1, 3), (1, 3)),
+            (("target t", "step"), (17, 0)),
+            (("target t", 0), (17, 0)),
+        ),
+        ids=(
+            "str",
+            "int",
+            "slice",
+            "tuple[int]",
+            "tuple[str]",
+            "tuple[mixed]",
+        ),
+    )
+    def test_getitem(self, ind, expectedcol):
+        assert (self.data[ind] == self.data.data[:, expectedcol]).all()
+
+    def test_repr(self, capsys):
+        print(self.data)
+        std_out_err = capsys.readouterr()
+        assert std_out_err.err == ""
+        assert "index label units" in std_out_err.out
+        assert (
+            f"contains {self.data.columns} timeseries, each with {self.data.rows} elements"
+            in std_out_err.out
+        )