BUG: hash_pandas_object fails on array containing tuple #28969 (#30508)

* BUG: hash_pandas_object fails on array containing tuple #28969

Author: jbrockmendel

doc/source/whatsnew/v1.0.0.rst

@@ -973,6 +973,7 @@ Other
 - Bug in :meth:`Series.diff` where a boolean series would incorrectly raise a ``TypeError`` (:issue:`17294`)
 - :meth:`Series.append` will no longer raise a ``TypeError`` when passed a tuple of ``Series`` (:issue:`28410`)
 - Fix corrupted error message when calling ``pandas.libs._json.encode()`` on a 0d array (:issue:`18878`)
+- Bug in ``pd.core.util.hashing.hash_pandas_object`` where arrays containing tuples were incorrectly treated as non-hashable (:issue:`28969`)
 - Bug in :meth:`DataFrame.append` that raised ``IndexError`` when appending with empty list (:issue:`28769`)
 - Fix :class:`AbstractHolidayCalendar` to return correct results for
  years after 2030 (now goes up to 2200) (:issue:`27790`)

pandas/_libs/hashing.pyx

@@ -70,6 +70,12 @@ def hash_object_array(object[:] arr, object key, object encoding='utf8'):
  # null, stringify and encode
  data = <bytes>str(val).encode(encoding)
 
+ elif isinstance(val, tuple):
+ # GH#28969 we could have a tuple, but need to ensure that
+ # the tuple entries are themselves hashable before converting
+ # to str
+ hash(val)
+ data = <bytes>str(val).encode(encoding)
  else:
  raise TypeError(f"{val} of type {type(val)} is not a valid type "
  "for hashing, must be string or null")

pandas/core/util/hashing.py

@@ -85,11 +85,12 @@ def hash_pandas_object(
  if isinstance(obj, ABCMultiIndex):
  return Series(hash_tuples(obj, encoding, hash_key), dtype="uint64", copy=False)
 
- if isinstance(obj, ABCIndexClass):
+ elif isinstance(obj, ABCIndexClass):
  h = hash_array(obj.values, encoding, hash_key, categorize).astype(
  "uint64", copy=False
  )
  h = Series(h, index=obj, dtype="uint64", copy=False)
+
  elif isinstance(obj, ABCSeries):
  h = hash_array(obj.values, encoding, hash_key, categorize).astype(
  "uint64", copy=False

pandas/tests/util/test_hashing.py

@@ -353,3 +353,24 @@ def test_hash_collisions():
 
  result = hash_array(np.asarray(hashes, dtype=object), "utf8")
  tm.assert_numpy_array_equal(result, np.concatenate([expected1, expected2], axis=0))
+
+
+def test_hash_with_tuple():
+ # GH#28969 array containing a tuple raises on call to arr.astype(str)
+ # apparently a numpy bug github.com/numpy/numpy/issues/9441
+
+ df = pd.DataFrame({"data": [tuple("1"), tuple("2")]})
+ result = hash_pandas_object(df)
+ expected = pd.Series([10345501319357378243, 8331063931016360761], dtype=np.uint64)
+ tm.assert_series_equal(result, expected)
+
+ df2 = pd.DataFrame({"data": [tuple([1]), tuple([2])]})
+ result = hash_pandas_object(df2)
+ expected = pd.Series([9408946347443669104, 3278256261030523334], dtype=np.uint64)
+ tm.assert_series_equal(result, expected)
+
+ # require that the elements of such tuples are themselves hashable
+
+ df3 = pd.DataFrame({"data": [tuple([1, []]), tuple([2, {}])]})
+ with pytest.raises(TypeError, match="unhashable type: 'list'"):
+ hash_pandas_object(df3)