BUG: raise on non-hashable in __contains__ (#30902)

Author: jbrockmendel

pandas/_libs/index.pyx

@@ -72,9 +72,10 @@ cdef class IndexEngine:
         self.over_size_threshold = n >= _SIZE_CUTOFF
         self.clear_mapping()
 
-    def __contains__(self, object val):
+    def __contains__(self, val: object) -> bool:
+        # We assume before we get here:
+        #  - val is hashable
         self._ensure_mapping_populated()
-        hash(val)
         return val in self.mapping
 
     cpdef get_value(self, ndarray arr, object key, object tz=None):
@@ -415,7 +416,9 @@ cdef class DatetimeEngine(Int64Engine):
             raise TypeError(scalar)
         return scalar.value
 
-    def __contains__(self, object val):
+    def __contains__(self, val: object) -> bool:
+        # We assume before we get here:
+        #  - val is hashable
         cdef:
             int64_t loc, conv
 
@@ -712,7 +715,9 @@ cdef class BaseMultiIndexCodesEngine:
 
         return indexer
 
-    def __contains__(self, object val):
+    def __contains__(self, val: object) -> bool:
+        # We assume before we get here:
+        #  - val is hashable
         # Default __contains__ looks in the underlying mapping, which in this
         # case only contains integer representations.
         try:

pandas/core/indexes/base.py

@@ -1,7 +1,7 @@
 from datetime import datetime
 import operator
 from textwrap import dedent
-from typing import Dict, FrozenSet, Hashable, Optional, Union
+from typing import Any, Dict, FrozenSet, Hashable, Optional, Union
 import warnings
 
 import numpy as np
@@ -4144,7 +4144,7 @@ def is_type_compatible(self, kind) -> bool:
         """
 
     @Appender(_index_shared_docs["contains"] % _index_doc_kwargs)
-    def __contains__(self, key) -> bool:
+    def __contains__(self, key: Any) -> bool:
         hash(key)
         try:
             return key in self._engine

pandas/core/indexes/category.py

@@ -385,11 +385,12 @@ def _wrap_setop_result(self, other, result):
         return self._shallow_copy(result, name=name)
 
     @Appender(_index_shared_docs["contains"] % _index_doc_kwargs)
-    def __contains__(self, key) -> bool:
+    def __contains__(self, key: Any) -> bool:
         # if key is a NaN, check if any NaN is in self.
         if is_scalar(key) and isna(key):
             return self.hasnans
 
+        hash(key)
         return contains(self, key, container=self._engine)
 
     def __array__(self, dtype=None) -> np.ndarray:

pandas/core/indexes/datetimelike.py

@@ -2,7 +2,7 @@
 Base and utility classes for tseries type pandas objects.
 """
 import operator
-from typing import List, Optional, Set
+from typing import Any, List, Optional, Set
 
 import numpy as np
 
@@ -154,7 +154,8 @@ def equals(self, other) -> bool:
         return np.array_equal(self.asi8, other.asi8)
 
     @Appender(_index_shared_docs["contains"] % _index_doc_kwargs)
-    def __contains__(self, key):
+    def __contains__(self, key: Any) -> bool:
+        hash(key)
         try:
             res = self.get_loc(key)
         except (KeyError, TypeError, ValueError):

pandas/core/indexes/interval.py

@@ -374,7 +374,7 @@ def _engine(self):
         right = self._maybe_convert_i8(self.right)
         return IntervalTree(left, right, closed=self.closed)
 
-    def __contains__(self, key) -> bool:
+    def __contains__(self, key: Any) -> bool:
         """
         return a boolean if this key is IN the index
         We *only* accept an Interval
@@ -387,6 +387,7 @@ def __contains__(self, key) -> bool:
         -------
         bool
         """
+        hash(key)
         if not isinstance(key, Interval):
             return False
 

pandas/core/indexes/multi.py

@@ -1,6 +1,6 @@
 import datetime
 from sys import getsizeof
-from typing import Hashable, List, Optional, Sequence, Union
+from typing import Any, Hashable, List, Optional, Sequence, Union
 import warnings
 
 import numpy as np
@@ -973,7 +973,7 @@ def _shallow_copy_with_infer(self, values, **kwargs):
         return self._shallow_copy(values, **kwargs)
 
     @Appender(_index_shared_docs["contains"] % _index_doc_kwargs)
-    def __contains__(self, key) -> bool:
+    def __contains__(self, key: Any) -> bool:
         hash(key)
         try:
             self.get_loc(key)

pandas/core/indexes/numeric.py

@@ -1,4 +1,4 @@
-from typing import TYPE_CHECKING
+from typing import TYPE_CHECKING, Any
 
 import numpy as np
 
@@ -461,7 +461,8 @@ def equals(self, other) -> bool:
         except (TypeError, ValueError):
             return False
 
-    def __contains__(self, other) -> bool:
+    def __contains__(self, other: Any) -> bool:
+        hash(other)
         if super().__contains__(other):
             return True
 

pandas/core/indexes/period.py

@@ -1,4 +1,5 @@
 from datetime import datetime, timedelta
+from typing import Any
 import weakref
 
 import numpy as np
@@ -358,18 +359,18 @@ def _engine(self):
         return self._engine_type(period, len(self))
 
     @Appender(_index_shared_docs["contains"])
-    def __contains__(self, key) -> bool:
+    def __contains__(self, key: Any) -> bool:
         if isinstance(key, Period):
             if key.freq != self.freq:
                 return False
             else:
                 return key.ordinal in self._engine
         else:
+            hash(key)
             try:
                 self.get_loc(key)
                 return True
-            except (TypeError, KeyError):
-                # TypeError can be reached if we pass a tuple that is not hashable
+            except KeyError:
                 return False
 
     @cache_readonly

pandas/core/indexes/range.py

@@ -1,7 +1,7 @@
 from datetime import timedelta
 import operator
 from sys import getsizeof
-from typing import Optional, Union
+from typing import Any, Optional
 import warnings
 
 import numpy as np
@@ -332,7 +332,7 @@ def is_monotonic_decreasing(self) -> bool:
     def has_duplicates(self) -> bool:
         return False
 
-    def __contains__(self, key: Union[int, np.integer]) -> bool:
+    def __contains__(self, key: Any) -> bool:
         hash(key)
         try:
             key = ensure_python_int(key)

pandas/tests/indexes/common.py

@@ -883,3 +883,11 @@ def test_getitem_2d_deprecated(self):
             res = idx[:, None]
 
         assert isinstance(res, np.ndarray), type(res)
+
+    def test_contains_requires_hashable_raises(self):
+        idx = self.create_index()
+        with pytest.raises(TypeError, match="unhashable type"):
+            [] in idx
+
+        with pytest.raises(TypeError):
+            {} in idx._engine