add whatsnew and make required changes

Author: Jiang Yue

doc/source/whatsnew/v0.25.0.rst

@@ -825,6 +825,7 @@ Other API changes
 - :meth:`ExtensionArray.argsort` places NA values at the end of the sorted array. (:issue:`21801`)
 - :meth:`DataFrame.to_hdf` and :meth:`Series.to_hdf` will now raise a ``NotImplementedError`` when saving a :class:`MultiIndex` with extention data types for a ``fixed`` format. (:issue:`7775`)
 - Passing duplicate ``names`` in :meth:`read_csv` will now raise a ``ValueError`` (:issue:`17346`)
+- :meth:`infer_type` will now return integer-na for integer and np.nan mix (:issue:`27283`)
 
 .. _whatsnew_0250.deprecations:
 

pandas/_libs/lib.pyx

@@ -1519,10 +1519,8 @@ cpdef bint is_integer_array(ndarray values):
 
 cdef class IntegerNaValidator(Validator):
     cdef inline bint is_value_typed(self, object value) except -1:
-        return util.is_integer_object(value) or util.is_nan(value)
-
-    cdef inline bint is_array_typed(self) except -1:
-        return issubclass(self.dtype.type, np.integer)
+        return util.is_integer_object(value) or (util.is_nan(value)
+                and util.is_float_object(value))
 
 
 cdef bint is_integer_na_array(ndarray values):

pandas/core/indexes/base.py

@@ -473,6 +473,7 @@ def __new__(
 
                     return Index(subarr, copy=copy, dtype=object, name=name)
                 elif inferred in ["floating", "mixed-integer-float", "integer-na"]:
+                    # TODO: Returns IntegerArray for integer-na case in the future
                     from .numeric import Float64Index
 
                     return Float64Index(subarr, copy=copy, name=name)
@@ -3197,10 +3198,7 @@ def is_int(v):
                     self.get_loc(stop)
                 is_positional = False
         except KeyError:
-            if (
-                self.inferred_type == "mixed-integer-float"
-                or self.inferred_type == "integer-na"
-            ):
+            if self.inferred_type in ["mixed-integer-float", "integer-na"]:
                 raise
 
         if is_null_slicer:

pandas/tests/dtypes/test_inference.py

@@ -577,15 +577,17 @@ def test_integers(self):
         result = lib.infer_dtype(arr, skipna=True)
         assert result == "integer"
 
-    # GH 27392
-    def test_integer_na(self):
+    @pytest.mark.parametrize("skipna", [True, False])
+    def test_integer_na(self, skipna):
+        # GH 27392
+        expected = "integer" if skipna else "integer-na"
         arr = np.array([1, 2, np.nan, np.nan, 3], dtype="O")
-        result = lib.infer_dtype(arr, skipna=False)
-        assert result == "integer-na"
+        result = lib.infer_dtype(arr, skipna=skipna)
+        assert result == expected
 
         arr = np.array([1, 2, 3, np.int64(4), np.int32(5), np.nan], dtype="O")
-        result = lib.infer_dtype(arr, skipna=False)
-        assert result == "integer-na"
+        result = lib.infer_dtype(arr, skipna=skipna)
+        assert result == expected
 
     def test_deprecation(self):
         # GH 24050