[Python] Test timestamp from int without pandas dependency

Author: HyukjinKwon

python/pyarrow/tests/test_convert_builtin.py

@@ -1381,12 +1381,8 @@ def test_sequence_timestamp_nanoseconds():
                                                       23, 34, 123456)
 
 
-@pytest.mark.pandas
 @pytest.mark.timezone_data
 def test_sequence_timestamp_from_int_with_unit():
-    # TODO(wesm): This test might be rewritten to assert the actual behavior
-    # when pandas is not installed
-
     data = [1]
 
     s = pa.timestamp('s')
@@ -1418,13 +1414,16 @@ def test_sequence_timestamp_from_int_with_unit():
     )
     assert str(arr_us[0]) == "1970-01-01 00:00:00.000001"
 
+    # Without pandas, nanosecond timestamps raise ValueError if value
+    # is not safely convertible to microseconds (value % 1000 != 0)
     arr_ns = pa.array(data, type=ns)
     assert len(arr_ns) == 1
     assert arr_ns.type == ns
-    assert repr(arr_ns[0].as_py()) == (
-        "Timestamp('1970-01-01 00:00:00.000000001')"
-    )
-    assert str(arr_ns[0]) == "1970-01-01 00:00:00.000000001"
+    try:
+        import pandas
+    except ImportError:
+        with pytest.raises(ValueError, match="not safely convertible to microseconds"):
+            arr_ns[0].as_py()
 
     expected_exc = TypeError
 
@@ -1436,6 +1435,28 @@ class CustomClass():
             pa.array([1, CustomClass()], type=ty)
 
 
+@pytest.mark.pandas
+@pytest.mark.timezone_data
+def test_sequence_timestamp_from_int_with_unit_nanosecond():
+    # With pandas installed, nanosecond timestamps return pd.Timestamp
+    # with full nanosecond precision (see scalar.pxi in _datetime_from_int)
+    import pandas as pd
+
+    data = [1]
+    ns = pa.timestamp('ns')
+
+    arr_ns = pa.array(data, type=ns)
+    assert len(arr_ns) == 1
+    assert arr_ns.type == ns
+
+    result = arr_ns[0].as_py()
+    assert isinstance(result, pd.Timestamp)
+    assert repr(result) == (
+        "Timestamp('1970-01-01 00:00:00.000000001')"
+    )
+    assert str(arr_ns[0]) == "1970-01-01 00:00:00.000000001"
+
+
 def test_sequence_duration():
     td1 = datetime.timedelta(2, 3601, 1)
     td2 = datetime.timedelta(1, 100, 1000)