Merge hgrecco#186

186: Add accumulate r=andrewgsavage a=andrewgsavage

- [x] Closes hgrecco#170
- [x] Executed `pre-commit run --all-files` with no errors
- [x] The change is fully covered by automated unit tests
- [x] Documented in docs/ as appropriate
- [x] Added an entry to the CHANGES file


Co-authored-by: Andrew <[email protected]>
Co-authored-by: andrewgsavage <[email protected]>

Author: bors[bot]

.github/workflows/ci-pint-master.yml

@@ -8,8 +8,8 @@ jobs:
       matrix:
         python-version: [3.9, "3.10", "3.11"]
         numpy: ["numpy>=1.20.3,<2.0.0"]
-        pandas: ["pandas==1.5.2", ]
-        pint: ["pint>=0.20.1"]
+        pandas: ["pandas==2.0.2", ]
+        pint: ["pint>=0.21.1"]
 
     runs-on: ubuntu-latest
 

.github/workflows/ci-pint-pre.yml

@@ -8,8 +8,8 @@ jobs:
       matrix:
         python-version: [3.9, "3.10", "3.11"]
         numpy: ["numpy>=1.20.3,<2.0.0"]
-        pandas: ["pandas==1.5.2", ]
-        pint: ["pint>=0.20.1"]
+        pandas: ["pandas==2.0.2", ]
+        pint: ["pint>=0.21.1"]
 
     runs-on: ubuntu-latest
 

.github/workflows/ci.yml

@@ -8,8 +8,8 @@ jobs:
       matrix:
         python-version: [3.9, "3.10", "3.11"]
         numpy: ["numpy>=1.20.3,<2.0.0"]
-        pandas: ["pandas==1.5.2","pandas==2.0.2", ]
-        pint: ["pint==0.21.1", "pint==0.22"]
+        pandas: ["pandas==2.0.2", ]
+        pint: ["pint>=0.21.1", "pint==0.22"]
 
     runs-on: ubuntu-latest
 

CHANGES

@@ -1,14 +1,23 @@
 pint-pandas Changelog
 =====================
 
-0.4 (unreleased)
+0.5 (unreleased)
+----------------
+
+- Support for pandas 2.0, allowing `.cumsum, .cummax, .cummin` methods for `Series` and `DataFrame`. #186
+- Minimum Pint version is 0.21
+- Minimum Pandas vesrion is 2.0
+- Support for unit registries with `force_ndarray_like = True`. #165
+
+0.4 (2023-05-23)
 ----------------
 
 - Support for <NA> values in columns with integer magnitudes
 - Support for magnitudes of any type, such as complex128 or tuples #146
 - Support for Pint 0.21 #168, #179
 - Cast to `numpy.ndarray` in `PintArray._reduce` if needed to use `nanops` functions
-- Support for unit registries with `force_ndarray_like = True`. #165
+- Minimum Pint version is 0.21
+- Minimum Pandas vesrion is 1.6
 
 0.3 (2022-11-14)
 ----------------

pint_pandas/pint_array.py

@@ -852,6 +852,24 @@ def _reduce(self, name, **kwds):
             return self._Q(result, self.units**2)
         return self._Q(result, self.units)
 
+    def _accumulate(self, name: str, *, skipna: bool = True, **kwds):
+        if name == "cumprod":
+            raise TypeError("cumprod not supported for pint arrays")
+        functions = {
+            "cummin": np.minimum.accumulate,
+            "cummax": np.maximum.accumulate,
+            "cumsum": np.cumsum,
+        }
+
+        if isinstance(self._data, ExtensionArray):
+            try:
+                result = self._data._accumulate(name, **kwds)
+            except NotImplementedError:
+                result = functions[name](self.numpy_data, **kwds)
+        print(result)
+
+        return self._from_sequence(result, self.units)
+
 
 PintArray._add_arithmetic_ops()
 PintArray._add_comparison_ops()

pint_pandas/testsuite/test_pandas_extensiontests.py

@@ -220,6 +220,17 @@ def all_boolean_reductions(request):
     return request.param
 
 
+_all_numeric_accumulations = ["cumsum", "cumprod", "cummin", "cummax"]
+
+
+@pytest.fixture(params=_all_numeric_accumulations)
+def all_numeric_accumulations(request):
+    """
+    Fixture for numeric accumulation names
+    """
+    return request.param
+
+
 @pytest.fixture
 def invalid_scalar(data):
     """
@@ -497,3 +508,20 @@ class TestSetitem(base.BaseSetitemTests):
     @pytest.mark.parametrize("numeric_dtype", _base_numeric_dtypes, indirect=True)
     def test_setitem_scalar_key_sequence_raise(self, data):
         base.BaseSetitemTests.test_setitem_scalar_key_sequence_raise(self, data)
+
+
+class TestAccumulate(base.BaseAccumulateTests):
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_accumulate_series_raises(self, data, all_numeric_accumulations, skipna):
+        pass
+
+    def check_accumulate(self, s, op_name, skipna):
+        if op_name == "cumprod":
+            with pytest.raises(TypeError):
+                getattr(s, op_name)(skipna=skipna)
+        else:
+            result = getattr(s, op_name)(skipna=skipna)
+            s_unitless = pd.Series(s.values.data)
+            expected = getattr(s_unitless, op_name)(skipna=skipna)
+            expected = pd.Series(expected, dtype=s.dtype)
+            self.assert_series_equal(result, expected, check_dtype=False)