Addressed review comments

pandas/core/algorithms.py

@@ -2061,32 +2061,32 @@ def sort_tuples(values):
  def cmp_func(index_x, index_y):
  x = values[index_x]
  y = values[index_y]
+ # shortcut loop in case both tuples are the same
  if x == y:
  return 0
- len_x = len(x)
- len_y = len(y)
- for i in range(max(len_x, len_y)):
+ # lexicographic sorting
+ for i in range(max(len(x), len(y))):
  # check if the tuples have different lengths (shorter tuples
  # first)
- if i >= len_x:
+ if i >= len(x):
  return -1
- if i >= len_y:
+ if i >= len(y):
  return +1
- x_i_na = isna(x[i])
- y_i_na = isna(y[i])
+ x_is_na = isna(x[i])
+ y_is_na = isna(y[i])
  # values are the same -> resolve tie with next element
- if (x_i_na and y_i_na) or (x[i] == y[i]):
+ if (x_is_na and y_is_na) or (x[i] == y[i]):
  continue
- # check for nan values (sort nan to the end which is consistent
- # with numpy
- if x_i_na and not y_i_na:
+ # check for nan values (sort nan to the end)
+ if x_is_na and not y_is_na:
  return +1
- if not x_i_na and y_i_na:
+ if not x_is_na and y_is_na:
  return -1
  # normal greater/less than comparison
  if x[i] < y[i]:
  return -1
  return +1
+ # both values are the same (should already have been caught)
  return 0
 
  ixs = np.arange(len(values))
@@ -2095,12 +2095,10 @@ def cmp_func(index_x, index_y):
 
  sorter = None
 
- ext_arr = is_extension_array_dtype(values)
- if not ext_arr and lib.infer_dtype(values, skipna=False) == "mixed-integer":
- # unorderable in py3 if mixed str/int
+ is_ea = is_extension_array_dtype(values)
+ if not is_ea and lib.infer_dtype(values, skipna=False) == "mixed-integer":
  ordered = sort_mixed(values)
- elif not ext_arr and values.size and isinstance(values[0], tuple):
- # 1-D arrays with tuples of potentially mixed type (solves GH36562)
+ elif not is_ea and values.size and isinstance(values[0], tuple):
  ordered = sort_tuples(values)
  else:
  try:

pandas/tests/indexing/multiindex/test_multiindex.py

@@ -93,7 +93,7 @@ def test_multiindex_get_loc_list_raises(self):
  idx.get_loc([])
 
 
-def test_combine_first_with_nan_index():
+def test_combine_first_with_nan_multiindex():
  mi1 = pd.MultiIndex.from_arrays(
  [["b", "b", "c", "a", "b", np.nan], [1, 2, 3, 4, 5, 6]], names=["a", "b"]
  )
@@ -102,17 +102,19 @@ def test_combine_first_with_nan_index():
  [["a", "b", "c", "a", "b", "d"], [1, 1, 1, 1, 1, 1]], names=["a", "b"]
  )
  s = pd.Series([1, 2, 3, 4, 5, 6], index=mi2)
- df_combined = df.combine_first(pd.DataFrame({"col": s}))
+ res = df.combine_first(pd.DataFrame({"d": s}))
  mi_expected = pd.MultiIndex.from_arrays(
  [
  ["a", "a", "a", "b", "b", "b", "b", "c", "c", "d", np.nan],
  [1, 1, 4, 1, 1, 2, 5, 1, 3, 1, 6],
  ],
  names=["a", "b"],
  )
- assert (df_combined.index == mi_expected).all()
- exp_col = np.asarray(
- [1.0, 4.0, np.nan, 2.0, 5.0, np.nan, np.nan, 3.0, np.nan, 6.0, np.nan]
+ expected = pd.DataFrame(
+ {
+ "c": [np.nan, np.nan, 1, 1, 1, 1, 1, np.nan, 1, np.nan, 1],
+ "d": [1.0, 4.0, np.nan, 2.0, 5.0, np.nan, np.nan, 3.0, np.nan, 6.0, np.nan],
+ },
+ index=mi_expected,
  )
- act_col = df_combined["col"].values
- assert np.allclose(act_col, exp_col, rtol=0, atol=0, equal_nan=True)
+ tm.assert_frame_equal(res, expected)