DOC, ENH, TST: Add more tests for duplicated()

Updates documentation to describe the "values"
parameter in the signature.

Adds functionality and tests for the following:

1) Accepting array-likes with no dtype attribute
2) Properly checking for duplicates in np.uint64

The algorithm currently will convert uint64 to int64,
but that is not an issue since all elements in the
uint64 range are unique in int64 as well. Hence,
the returned result is ultimately correct.

Author: gfyoung

pandas/core/algorithms.py

@@ -490,12 +490,14 @@ def _value_counts_arraylike(values, dropna=True):
 
 def duplicated(values, keep='first'):
     """
-    Return boolean ndarray denoting duplicate values
+    Return boolean ndarray denoting duplicate values.
 
     .. versionadded:: 0.19.0
 
     Parameters
     ----------
+    values : array-like
+        Array over which to check for duplicate values.
     keep : {'first', 'last', False}, default 'first'
         - ``first`` : Mark duplicates as ``True`` except for the first
           occurrence.
@@ -508,6 +510,7 @@ def duplicated(values, keep='first'):
     duplicated : ndarray
     """
 
+    values = com._asarray_tuplesafe(values)
     dtype = values.dtype
 
     # no need to revert to original type
@@ -522,6 +525,8 @@ def duplicated(values, keep='first'):
         values = values.values
 
     if is_integer_dtype(dtype):
+        # This also works if dtype is uint64 because there is a 1-1
+        # correspondence between int64 and uint64.
         values = _ensure_int64(values)
         duplicated = htable.duplicated_int64(values, keep=keep)
     elif is_float_dtype(dtype):

pandas/tests/test_algos.py

@@ -672,7 +672,9 @@ def test_numeric_object_likes(self):
                  np.array([1 + 1j, 2 + 2j, 1 + 1j, 5 + 5j, 3 + 3j,
                            2 + 2j, 4 + 4j, 1 + 1j, 5 + 5j, 6 + 6j]),
                  np.array(['a', 'b', 'a', 'e', 'c',
-                           'b', 'd', 'a', 'e', 'f'], dtype=object)]
+                           'b', 'd', 'a', 'e', 'f'], dtype=object),
+                 np.array([1, 2**63, 1, 3**5, 10,
+                           2**63, 39, 1, 3**5, 7], dtype=np.uint64)]
 
         exp_first = np.array([False, False, True, False, False,
                               True, False, True, True, False])
@@ -772,6 +774,13 @@ def test_unique_index(self):
             tm.assert_numpy_array_equal(case.duplicated(),
                                         np.array([False, False, False]))
 
+    def test_duplicated_non_dtype(self):
+        # Make sure we can pass in array-likes with no dtype attribute.
+        cases = [[1, 2, 3], tuple([1, 2, 3])]
+        for case in cases:
+            tm.assert_numpy_array_equal(algos.duplicated(case),
+                                        np.array([False, False, False]))
+
 
 class GroupVarTestMixin(object):