added specific test

Author: Tom Augspurger

doc/source/whatsnew/v1.4.0.rst

@@ -203,6 +203,7 @@ Other enhancements
 - :meth:`IntegerArray.all` , :meth:`IntegerArray.any`, :meth:`FloatingArray.any`, and :meth:`FloatingArray.all` use Kleene logic (:issue:`41967`)
 - Added support for nullable boolean and integer types in :meth:`DataFrame.to_stata`, :class:`~pandas.io.stata.StataWriter`, :class:`~pandas.io.stata.StataWriter117`, and :class:`~pandas.io.stata.StataWriterUTF8` (:issue:`40855`)
 - :meth:`DataFrame.__pos__`, :meth:`DataFrame.__neg__` now retain ``ExtensionDtype`` dtypes (:issue:`43883`)
+- Added :meth:`api.extension.ExtensionArray._format_array` for extension arrays to control how they are formatted in ``Series`` and ``DataFrame`` (:issue:`26837`)
 - The error raised when an optional dependency can't be imported now includes the original exception, for easier investigation (:issue:`43882`)
 - Added :meth:`.ExponentialMovingWindow.sum` (:issue:`13297`)
 - :meth:`Series.str.split` now supports a ``regex`` argument that explicitly specifies whether the pattern is a regular expression. Default is ``None`` (:issue:`43563`, :issue:`32835`, :issue:`25549`)

pandas/tests/extension/test_format.py

@@ -0,0 +1,44 @@
+import pandas as pd
+
+
+class MyDtype(pd.api.extensions.ExtensionDtype):
+    name = "mydtype"
+    type = list
+
+
+class MyEA(pd.api.extensions.ExtensionArray):
+    def __init__(self, data):
+        self.data = data
+        self._dtype = MyDtype()
+
+    @property
+    def dtype(self):
+        return self._dtype
+
+    def __len__(self):
+        return 1
+
+    def __array__(self, dtype=None):
+        raise ValueError("Cannot be converted to an array!")
+
+    def _format_array(
+        self,
+        formatter: None,
+        float_format: None,
+        na_rep="NaN",
+        digits=None,
+        space=None,
+        justify="right",
+        decimal=".",
+        leading_space=True,
+        quoting=None,
+    ):
+        return ["<MyEA>([1])"]
+
+
+def test_no_conversion():
+    s = pd.Series(MyEA([1]))
+    repr(s)  # OK!
+
+    df = pd.DataFrame({"A": MyEA([1])}, copy=False)
+    repr(df)  # OK!