From b4a48d0d48955070ec9e73a8a85c255f538a209a Mon Sep 17 00:00:00 2001 From: Hyukjin Kwon Date: Mon, 13 May 2024 11:42:17 +0900 Subject: [PATCH] Fix nested array to respect legacy conf of inferArrayTypeFromFirstElement --- python/pyspark/sql/tests/test_types.py | 7 +++++++ python/pyspark/sql/types.py | 18 ++++++++++++++++-- 2 files changed, 23 insertions(+), 2 deletions(-) diff --git a/python/pyspark/sql/tests/test_types.py b/python/pyspark/sql/tests/test_types.py index 40eded6a4433c..1596789376834 100644 --- a/python/pyspark/sql/tests/test_types.py +++ b/python/pyspark/sql/tests/test_types.py @@ -1621,6 +1621,13 @@ def test_collated_string(self): StringType("UTF8_BINARY_LCASE"), ) + def test_infer_array_element_type_with_struct(self): + # SPARK-48248: Nested array to respect legacy conf of inferArrayTypeFromFirstElement + with self.sql_conf( + {"spark.sql.pyspark.legacy.inferArrayTypeFromFirstElement.enabled": True} + ): + self.assertEqual([[1, None]], self.spark.createDataFrame([[[[1, "a"]]]]).first()[0]) + class DataTypeTests(unittest.TestCase): # regression test for SPARK-6055 diff --git a/python/pyspark/sql/types.py b/python/pyspark/sql/types.py index 41be12620fd56..fbd4987713e26 100644 --- a/python/pyspark/sql/types.py +++ b/python/pyspark/sql/types.py @@ -1951,13 +1951,27 @@ def _infer_type( if len(obj) > 0: if infer_array_from_first_element: return ArrayType( - _infer_type(obj[0], infer_dict_as_struct, prefer_timestamp_ntz), True + _infer_type( + obj[0], + infer_dict_as_struct, + infer_array_from_first_element, + prefer_timestamp_ntz, + ), + True, ) else: return ArrayType( reduce( _merge_type, - (_infer_type(v, infer_dict_as_struct, prefer_timestamp_ntz) for v in obj), + ( + _infer_type( + v, + infer_dict_as_struct, + infer_array_from_first_element, + prefer_timestamp_ntz, + ) + for v in obj + ), ), True, )