Merge hgrecco#58

58: [WIP] Convert object dtype r=andrewgsavage a=andrewgsavage

Addresses hgrecco#55

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

Author: bors[bot]

CHANGES

@@ -4,10 +4,14 @@ pint-pandas Changelog
 0.5 (unreleased)
 ----------------
 
+<<<<<<< HEAD
+- Support for <NA> values in columns with integer magnitudes
+- Support for magnitudes of any type, such as complex128 or tuples #146
 - 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
+- A DataFrame/Series.pint.convert_object_dtype() function has been added to create PintArrays from Series of quantities.
 
 0.4 (2023-05-23)
 ----------------

pint_pandas/pint_array.py

@@ -948,15 +948,32 @@ def to_base_units(self):
             index=index,
         )
 
+    def convert_object_dtype(self):
+        df = self._obj
+        df_new = pd.DataFrame()
+        for col in df.columns:
+            s = df[col]
+            if s.dtype == "object":
+                try:
+                    df_new[col] = s.pint.convert_object_dtype()
+                except AttributeError:
+                    df_new[col] = s
+            else:
+                df_new[col] = s
+        return df_new
+
 
 @register_series_accessor("pint")
 class PintSeriesAccessor(object):
     def __init__(self, pandas_obj):
-        self._validate(pandas_obj)
-        self.pandas_obj = pandas_obj
-        self.quantity = pandas_obj.values.quantity
-        self._index = pandas_obj.index
-        self._name = pandas_obj.name
+        if self._is_object_dtype_and_quantity(pandas_obj):
+            self.pandas_obj = pandas_obj
+        else:
+            self._validate(pandas_obj)
+            self.pandas_obj = pandas_obj
+            self.quantity = pandas_obj.values.quantity
+            self._index = pandas_obj.index
+            self._name = pandas_obj.name
 
     @staticmethod
     def _validate(obj):
@@ -966,6 +983,19 @@ def _validate(obj):
                 "dtype '{}'.".format(obj.dtype)
             )
 
+    @staticmethod
+    def _is_object_dtype_and_quantity(obj):
+        return obj.dtype == "object" and all(
+            [(isinstance(item, _Quantity) or pd.isna(item)) for item in obj.values]
+        )
+
+    def convert_object_dtype(self):
+        return pd.Series(
+            data=PintArray._from_sequence(self.pandas_obj.values),
+            index=self.pandas_obj.index,
+            name=self.pandas_obj.name,
+        )
+
 
 class Delegated:
     # Descriptor for delegating attribute access to from

pint_pandas/testsuite/test_pandas_interface.py

@@ -310,6 +310,16 @@ def test_series_method_accessors(self, data, attr_args):
         s = pd.Series(data)
         assert all(getattr(s.pint, attr)(*args) == getattr(data.quantity, attr)(*args))
 
+    def test_convert_object_dtype(self, data):
+        ser = pd.Series(data)
+        ser_obj = pd.Series(ser.values, dtype="object")
+        assert ser_obj.pint.convert_object_dtype().dtype == ser.dtype
+
+        df = pd.DataFrame({"A": ser, "B": ser})
+        df2 = pd.DataFrame({"A": ser, "B": ser_obj})
+
+        assert all(df2.pint.convert_object_dtype().dtypes == df.dtypes)
+
 
 arithmetic_ops = [
     operator.add,