[backport 2.3.x] String dtype: more informative repr (keeping brief __str__) (#61148) (#62329)

Author: jorisvandenbossche

pandas/core/arrays/string_.py

@@ -119,10 +119,10 @@ class StringDtype(StorageExtensionDtype):
     Examples
     --------
     >>> pd.StringDtype()
-    string[python]
+    <StringDtype(storage='python', na_value=<NA>)>
 
     >>> pd.StringDtype(storage="pyarrow")
-    string[pyarrow]
+    <StringDtype(na_value=<NA>)>
     """
 
     @property
@@ -194,11 +194,8 @@ def __init__(
         self._na_value = na_value
 
     def __repr__(self) -> str:
-        if self._na_value is libmissing.NA:
-            return f"{self.name}[{self.storage}]"
-        else:
-            # TODO add more informative repr
-            return self.name
+        storage = "" if self.storage == "pyarrow" else "storage='python', "
+        return f"<StringDtype({storage}na_value={self._na_value})>"
 
     def __eq__(self, other: object) -> bool:
         # we need to override the base class __eq__ because na_value (NA or NaN)

pandas/core/generic.py

@@ -7021,12 +7021,12 @@ def convert_dtypes(
         2  3  z   <NA>  <NA>    20  200.0
 
         >>> dfn.dtypes
-        a             Int32
-        b    string[python]
-        c           boolean
-        d    string[python]
-        e             Int64
-        f           Float64
+        a      Int32
+        b     string
+        c    boolean
+        d     string
+        e      Int64
+        f    Float64
         dtype: object
 
         Start with a Series of strings and missing data represented by ``np.nan``.

pandas/io/formats/format.py

@@ -66,7 +66,6 @@
     ExtensionArray,
     TimedeltaArray,
 )
-from pandas.core.arrays.string_ import StringDtype
 from pandas.core.base import PandasObject
 import pandas.core.common as com
 from pandas.core.indexes.api import (
@@ -1232,8 +1231,6 @@ def _format(x):
                 return self.na_rep
             elif isinstance(x, PandasObject):
                 return str(x)
-            elif isinstance(x, StringDtype):
-                return repr(x)
             else:
                 # object dtype
                 return str(formatter(x))

pandas/tests/arrays/string_/test_string.py

@@ -123,6 +123,18 @@ def test_repr(dtype):
     assert repr(df.A.array) == expected
 
 
+def test_dtype_repr(dtype):
+    if dtype.storage == "pyarrow":
+        if dtype.na_value is pd.NA:
+            assert repr(dtype) == "<StringDtype(na_value=<NA>)>"
+        else:
+            assert repr(dtype) == "<StringDtype(na_value=nan)>"
+    elif dtype.na_value is pd.NA:
+        assert repr(dtype) == "<StringDtype(storage='python', na_value=<NA>)>"
+    else:
+        assert repr(dtype) == "<StringDtype(storage='python', na_value=nan)>"
+
+
 def test_none_to_nan(cls, dtype):
     a = cls._from_sequence(["a", None, "b"], dtype=dtype)
     assert a[1] is not None

pandas/tests/io/formats/test_to_string.py

@@ -758,9 +758,9 @@ def test_to_string_string_dtype(self):
         result = df.dtypes.to_string()
         expected = dedent(
             """\
-            x    string[pyarrow]
-            y     string[python]
-            z     int64[pyarrow]"""
+            x            string
+            y            string
+            z    int64[pyarrow]"""
         )
         assert result == expected