Fix support for pandas.CategoricalDtype (#205)

This regressed in `2.8.1`.

Fix #197

Author: nicoddemus

CHANGELOG.rst

@@ -1,3 +1,10 @@
+UNRELEASED
+----------
+
+*UNRELEASED*
+
+* `#197 <https://github.com/ESSS/pytest-regressions/issues/197>`__: Fixed support for ``pandas.CategoricalDtype``, regressed in ``2.8.1``.
+
 2.8.2
 -----
 

src/pytest_regressions/dataframe_regression.py

@@ -249,7 +249,12 @@ def check(
 
         for column in data_frame.columns:
             array = data_frame[column]
-            # Skip assertion if an array of strings
+
+            # Categorical columns (numbers mapped to strings) are supported.
+            if isinstance(array.dtype, pd.CategoricalDtype):
+                continue
+
+            # Arrays of strings are supported.
             if (array.dtype == "O") and (type(array.iloc[0]) is str):
                 continue
             # Rejected: timedelta, objects, zero-terminated bytes, unicode strings and raw data

tests/test_dataframe_regression.py

@@ -6,6 +6,7 @@
 import pandas as pd
 import pytest
 
+from pytest_regressions.dataframe_regression import DataFrameRegressionFixture
 from pytest_regressions.testing import check_regression_fixture_workflow
 
 
@@ -284,3 +285,15 @@ def test_dataframe_with_datetime(dataframe_regression):
     )
 
     dataframe_regression.check(df)
+
+
+def test_categorical(dataframe_regression: DataFrameRegressionFixture) -> None:
+    """Support for categorical data columns (#197)."""
+    values = [3, 3, 1, 1, 2, 0, 0, 2]
+    types = pd.Categorical.from_codes(
+        np.array(values),
+        categories=["a", "b", "c", "d"],
+    )
+
+    df = pd.DataFrame.from_dict({"types": types})
+    dataframe_regression.check(df)

tests/test_dataframe_regression/test_categorical.csv

@@ -0,0 +1,9 @@
+,types
+0,d
+1,d
+2,b
+3,b
+4,c
+5,a
+6,a
+7,c