Set seed=0 for some integration test cases

integration_tests/src/main/python/cast_test.py

@@ -205,6 +205,7 @@ def test_ansi_cast_decimal_to(data_gen, to_type):
             conf = {'spark.rapids.sql.castDecimalToFloat.enabled': True,
                 'spark.sql.ansi.enabled': True})
 
+@datagen_overrides(seed=0, reason='https://github.com/NVIDIA/spark-rapids/issues/10050')
 @pytest.mark.parametrize('data_gen', [
     DecimalGen(7, 1),
     DecimalGen(9, 9),

integration_tests/src/main/python/date_time_test.py

@@ -17,7 +17,7 @@
 from conftest import is_utc, is_supported_time_zone
 from data_gen import *
 from datetime import date, datetime, timezone
-from marks import ignore_order, incompat, allow_non_gpu
+from marks import ignore_order, incompat, allow_non_gpu, datagen_overrides
 from pyspark.sql.types import *
 from spark_session import with_cpu_session, is_before_spark_330, is_before_spark_350
 import pyspark.sql.functions as f
@@ -207,6 +207,7 @@ def test_datesub(data_gen):
 # than -106032829 for date('0001-01-01') so we have to cap the days values to the lower upper and
 # lower ranges.
 to_unix_timestamp_days_gen=[ByteGen(), ShortGen(), IntegerGen(min_val=-106032829, max_val=103819094, special_cases=[-106032829, 103819094,0,1,-1])]
+@datagen_overrides(seed=0, reason='https://github.com/NVIDIA/spark-rapids/issues/10027')
 @pytest.mark.parametrize('data_gen', to_unix_timestamp_days_gen, ids=idfn)
 @incompat
 @allow_non_gpu(*non_utc_allow)
@@ -221,6 +222,7 @@ def test_dateadd_with_date_overflow(data_gen):
            'unix_timestamp(date_add(a, cast(24 as {})))'.format(string_type)))
 
 to_unix_timestamp_days_gen=[ByteGen(), ShortGen(), IntegerGen(max_val=106032829, min_val=-103819094, special_cases=[106032829, -103819094,0,1,-1])]
+@datagen_overrides(seed=0, reason='https://github.com/NVIDIA/spark-rapids/issues/10027')
 @pytest.mark.parametrize('data_gen', to_unix_timestamp_days_gen, ids=idfn)
 @incompat
 @allow_non_gpu(*non_utc_allow)

integration_tests/src/main/python/delta_lake_update_test.py

@@ -142,6 +142,7 @@ def generate_dest_data(spark):
 @pytest.mark.parametrize("partition_columns", [None, ["a"]], ids=idfn)
 @pytest.mark.parametrize("enable_deletion_vectors", [True, False], ids=idfn)
 @pytest.mark.skipif(not supports_delta_lake_deletion_vectors(), reason="Deletion vectors are new in Spark 3.4.0 / DBR 12.2")
+@datagen_overrides(seed=0, reason='https://github.com/NVIDIA/spark-rapids/issues/10025')
 def test_delta_update_rows_with_dv(spark_tmp_path, use_cdf, partition_columns, enable_deletion_vectors):
     # Databricks changes the number of files being written, so we cannot compare logs unless there's only one slice
     num_slices_to_test = 1 if is_databricks_runtime() else 10
@@ -160,6 +161,7 @@ def generate_dest_data(spark):
 @pytest.mark.parametrize("use_cdf", [True, False], ids=idfn)
 @pytest.mark.parametrize("partition_columns", [None, ["a"]], ids=idfn)
 @pytest.mark.skipif(is_before_spark_320(), reason="Delta Lake writes are not supported before Spark 3.2.x")
+@datagen_overrides(seed=0, reason='https://github.com/NVIDIA/spark-rapids/issues/10025')
 def test_delta_update_dataframe_api(spark_tmp_path, use_cdf, partition_columns):
     from delta.tables import DeltaTable
     data_path = spark_tmp_path + "/DELTA_DATA"

integration_tests/src/main/python/hash_aggregate_test.py

@@ -1168,6 +1168,7 @@ def test_hash_multiple_filters(data_gen, conf):
         'min(a), max(b) filter (where c > 250) from hash_agg_table group by a',
         conf)
 
+@datagen_overrides(seed=0, reason='https://github.com/NVIDIA/spark-rapids/issues/10026')
 @approximate_float
 @ignore_order
 @pytest.mark.parametrize('data_gen', [_grpkey_floats_with_nan_zero_grouping_keys,

integration_tests/src/main/python/orc_cast_test.py

@@ -1,4 +1,4 @@
-# Copyright (c) 2020-2022, NVIDIA CORPORATION.
+# Copyright (c) 2020-2023, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -17,7 +17,7 @@
 from asserts import assert_gpu_and_cpu_are_equal_collect, assert_gpu_and_cpu_error
 from conftest import is_not_utc
 from data_gen import *
-from marks import allow_non_gpu
+from marks import allow_non_gpu, datagen_overrides
 from pyspark.sql.types import *
 from spark_session import with_cpu_session
 from orc_test import reader_opt_confs
@@ -103,6 +103,7 @@ def test_casting_from_float_and_double(spark_tmp_path, to_type):
     )
 
 
+@datagen_overrides(seed=0, reason='https://github.com/NVIDIA/spark-rapids/issues/10017')
 @pytest.mark.parametrize('data_gen', [DoubleGen(max_exp=32, special_cases=None),
                                       DoubleGen(max_exp=32, special_cases=[8.88e9, 9.99e10, 1.314e11])])
 @allow_non_gpu(*non_utc_allow_orc_scan)