parametrize test, allow max_seq_items to determine truncation

Author: Marco Gorelli

pandas/core/indexes/base.py

@@ -4766,7 +4766,7 @@ def _maybe_promote(self, other):
                 return self.astype("object"), other.astype("object")
         return self, other
 
-    def groupby(self, values) -> PrettyDict:
+    def groupby(self, values) -> PrettyDict[Hashable, np.ndarray]:
         """
         Group the index labels by a given array of values.
 

pandas/io/formats/printing.py

@@ -102,7 +102,7 @@ def _pprint_seq(
 ) -> str:
     """
     internal. pprinter for iterables. you should probably use pprint_thing()
-    rather than calling this directly.
+    rather then calling this directly.
 
     bounds length of printed sequence, depending on options
     """
@@ -137,7 +137,7 @@ def _pprint_dict(
 ) -> str:
     """
     internal. pprinter for iterables. you should probably use pprint_thing()
-    rather than calling this directly.
+    rather then calling this directly.
     """
     fmt = "{{{things}}}"
     pairs = []
@@ -538,4 +538,4 @@ class PrettyDict(Dict[_KT, _VT]):
     """Dict extension to support abbreviated __repr__"""
 
     def __repr__(self) -> str:
-        return pprint_thing(self, max_seq_items=get_option("display.max_rows"))
+        return pprint_thing(self)

pandas/tests/groupby/test_groupby.py

@@ -2039,14 +2039,21 @@ def test_groupby_list_level():
     tm.assert_frame_equal(result, expected)
 
 
-def test_groups_repr_truncates():
+@pytest.mark.parametrize(
+    "max_seq_items, expected",
+    [
+        (5, "{0: [0], 1: [1], 2: [2], 3: [3], 4: [4]}"),
+        (4, "{0: [0], 1: [1], 2: [2], 3: [3], ...}"),
+    ],
+)
+def test_groups_repr_truncates(max_seq_items, expected):
     # GH 1135
-    df = pd.DataFrame(np.random.randn(61, 1))
+    df = pd.DataFrame(np.random.randn(5, 1))
     df["a"] = df.index
 
-    result = df.groupby("a").groups.__repr__()
-    expected = str({i: [i] for i in range(60)})[:-1] + ", ...}"
-    assert result == expected
+    with pd.option_context("display.max_seq_items", max_seq_items):
+        result = df.groupby("a").groups.__repr__()
+        assert result == expected
 
-    result = df.groupby(np.array(df.a)).groups.__repr__()
-    assert result == expected
+        result = df.groupby(np.array(df.a)).groups.__repr__()
+        assert result == expected