TST: Add test for the large file read/write

python/pyarrow/tests/conftest.py

@@ -15,7 +15,7 @@
 # specific language governing permissions and limitations
 # under the License.
 
-from pytest import skip
+from pytest import skip, mark
 
 
 groups = [
@@ -70,6 +70,18 @@ def pytest_addoption(parser):
                          default=False,
                          help=('Run only the {0} test group'.format(group)))
 
+    parser.addoption('--runslow', action='store_true',
+                     default=False, help='run slow tests')
+
+
+def pytest_collection_modifyitems(config, items):
+    if not config.getoption('--runslow'):
+        skip_slow = mark.skip(reason='need --runslow option to run')
+
+        for item in items:
+            if 'slow' in item.keywords:
+                item.add_marker(skip_slow)
+
 
 def pytest_runtest_setup(item):
     only_set = False

python/pyarrow/tests/test_feather.py

@@ -50,7 +50,7 @@ def tearDown(self):
                 pass
 
     def test_file_not_exist(self):
-        with self.assertRaises(pa.ArrowIOError):
+        with pytest.raises(pa.ArrowIOError):
             FeatherReader('test_invalid_file')
 
     def _get_null_counts(self, path, columns=None):
@@ -98,7 +98,7 @@ def _assert_error_on_write(self, df, exc, path=None):
         def f():
             write_feather(df, path)
 
-        self.assertRaises(exc, f)
+        pytest.raises(exc, f)
 
     def test_num_rows_attr(self):
         df = pd.DataFrame({'foo': [1, 2, 3, 4, 5]})
@@ -466,3 +466,9 @@ def test_unsupported(self):
         # non-strings
         df = pd.DataFrame({'a': ['a', 1, 2.0]})
         self._assert_error_on_write(df, ValueError)
+
+    @pytest.mark.slow
+    def test_large_dataframe(self):
+        df = pd.DataFrame(np.random.randint(0, 100, size=(400000000, 1)),
+                          columns=list('A'))
+        self._check_pandas_roundtrip(df)