From e35cf5cd61dec5f50ff4831e45e5d9165e1831f9 Mon Sep 17 00:00:00 2001
From: Kaushik Srinivasan <kaushik.srinii@gmail.com>
Date: Thu, 20 Feb 2025 19:45:27 -0500
Subject: [PATCH] add integration test for pyarrow reads

---
 tests/integration/test_reads.py | 37 +++++++++++++++++++++++++++++++++
 1 file changed, 37 insertions(+)

diff --git a/tests/integration/test_reads.py b/tests/integration/test_reads.py
index ee5f8a2574..444091d944 100644
--- a/tests/integration/test_reads.py
+++ b/tests/integration/test_reads.py
@@ -56,6 +56,7 @@
     NestedField,
     StringType,
     TimestampType,
+    UnknownType,
 )
 from pyiceberg.utils.concurrent import ExecutorFactory
 
@@ -978,3 +979,39 @@ def test_scan_with_datetime(catalog: Catalog) -> None:
 
     df = table.scan(row_filter=LessThan("datetime", yesterday)).to_pandas()
     assert len(df) == 0
+
+
+@pytest.mark.integration
+@pytest.mark.parametrize("catalog", [pytest.lazy_fixture("session_catalog_hive")])
+def test_read_unknown_type(catalog: Catalog) -> None:
+    identifier = "default.test_table_read_unknown_type"
+    arrow_table = pa.Table.from_pydict(
+        {
+            "int": [1, 2],
+            "string": ["a", "b"],
+            "unknown": [None, None],
+        },
+        schema=pa.schema(
+            [
+                pa.field("int", pa.int32(), nullable=True),
+                pa.field("string", pa.string(), nullable=True),
+                pa.field("unknown", pa.null(), nullable=True),
+            ],
+        ),
+    )
+
+    try:
+        catalog.drop_table(identifier)
+    except NoSuchTableError:
+        pass
+
+    tbl = catalog.create_table(
+        identifier,
+        schema=arrow_table.schema,
+    )
+
+    tbl.append(arrow_table)
+
+    assert tbl.schema().find_type("unknown") == UnknownType()
+    result_table = tbl.scan().to_arrow()
+    assert result_table["unknown"].to_pylist() == [None, None]
\ No newline at end of file