Skip to content

Commit

Permalink
xfail failing tests due to WriteIntoDeltaCommand
Browse files Browse the repository at this point in the history
  • Loading branch information
@razajafri
razajafri committed Nov 12, 2023
1 parent 06a5770 commit 00b498e
Show file tree
Hide file tree
Showing 2 changed files with 11 additions and 2 deletions.
3 changes: 2 additions & 1 deletion integration_tests/src/main/python/delta_lake_update_test.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,7 @@
from delta_lake_utils import *
from marks import *
from spark_session import is_before_spark_320, is_databricks_runtime, \
supports_delta_lake_deletion_vectors, with_cpu_session, with_gpu_session
supports_delta_lake_deletion_vectors, with_cpu_session, with_gpu_session, is_spark_340_or_later

delta_update_enabled_conf = copy_and_update(delta_writes_enabled_conf,
{"spark.rapids.sql.command.UpdateCommand": "true",
Expand Down Expand Up @@ -71,6 +71,7 @@ def checker(data_path, do_update):
delta_writes_enabled_conf # Test disabled by default
], ids=idfn)
@pytest.mark.skipif(is_before_spark_320(), reason="Delta Lake writes are not supported before Spark 3.2.x")
@pytest.mark.xfail(condition=is_spark_340_or_later() and is_databricks_runtime(), reason="https://github.com/NVIDIA/spark-rapids/issues/9675")
def test_delta_update_disabled_fallback(spark_tmp_path, disable_conf):
data_path = spark_tmp_path + "/DELTA_DATA"
def setup_tables(spark):
Expand Down
10 changes: 9 additions & 1 deletion integration_tests/src/main/python/delta_lake_write_test.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,7 @@
from marks import *
from parquet_write_test import parquet_part_write_gens, parquet_write_gens_list, writer_confs
from pyspark.sql.types import *
from spark_session import is_before_spark_320, is_before_spark_330, is_spark_340_or_later, with_cpu_session
from spark_session import is_before_spark_320, is_before_spark_330, is_spark_340_or_later, with_cpu_session, is_databricks_runtime

delta_write_gens = [x for sublist in parquet_write_gens_list for x in sublist]

Expand Down Expand Up @@ -65,6 +65,7 @@ def do_sql(spark, q): spark.sql(q)
{"spark.rapids.sql.format.parquet.enabled": "false"},
{"spark.rapids.sql.format.parquet.write.enabled": "false"}], ids=idfn)
@pytest.mark.skipif(is_before_spark_320(), reason="Delta Lake writes are not supported before Spark 3.2.x")
@pytest.mark.xfail(condition=is_spark_340_or_later() and is_databricks_runtime(), reason="https://github.com/NVIDIA/spark-rapids/issues/9675")
def test_delta_write_disabled_fallback(spark_tmp_path, disable_conf):
data_path = spark_tmp_path + "/DELTA_DATA"
assert_gpu_fallback_write(
Expand Down Expand Up @@ -403,6 +404,7 @@ def setup_tables(spark):
@ignore_order
@pytest.mark.parametrize("ts_write", ["INT96", "TIMESTAMP_MICROS", "TIMESTAMP_MILLIS"], ids=idfn)
@pytest.mark.skipif(is_before_spark_320(), reason="Delta Lake writes are not supported before Spark 3.2.x")
@pytest.mark.xfail(condition=is_spark_340_or_later() and is_databricks_runtime(), reason="https://github.com/NVIDIA/spark-rapids/issues/9675")
def test_delta_write_legacy_timestamp_fallback(spark_tmp_path, ts_write):
gen = TimestampGen(start=datetime(1590, 1, 1, tzinfo=timezone.utc))
data_path = spark_tmp_path + "/DELTA_DATA"
Expand All @@ -425,6 +427,7 @@ def test_delta_write_legacy_timestamp_fallback(spark_tmp_path, ts_write):
{"parquet.encryption.column.keys": "k2:a"},
{"parquet.encryption.footer.key": "k1", "parquet.encryption.column.keys": "k2:a"}])
@pytest.mark.skipif(is_before_spark_320(), reason="Delta Lake writes are not supported before Spark 3.2.x")
@pytest.mark.xfail(condition=is_spark_340_or_later() and is_databricks_runtime(), reason="https://github.com/NVIDIA/spark-rapids/issues/9675")
def test_delta_write_encryption_option_fallback(spark_tmp_path, write_options):
def write_func(spark, path):
writer = unary_op_df(spark, int_gen).coalesce(1).write.format("delta")
Expand All @@ -446,6 +449,7 @@ def write_func(spark, path):
{"parquet.encryption.column.keys": "k2:a"},
{"parquet.encryption.footer.key": "k1", "parquet.encryption.column.keys": "k2:a"}])
@pytest.mark.skipif(is_before_spark_320(), reason="Delta Lake writes are not supported before Spark 3.2.x")
@pytest.mark.xfail(condition=is_spark_340_or_later() and is_databricks_runtime(), reason="https://github.com/NVIDIA/spark-rapids/issues/9675")
def test_delta_write_encryption_runtimeconfig_fallback(spark_tmp_path, write_options):
data_path = spark_tmp_path + "/DELTA_DATA"
assert_gpu_fallback_write(
Expand All @@ -462,6 +466,7 @@ def test_delta_write_encryption_runtimeconfig_fallback(spark_tmp_path, write_opt
{"parquet.encryption.column.keys": "k2:a"},
{"parquet.encryption.footer.key": "k1", "parquet.encryption.column.keys": "k2:a"}])
@pytest.mark.skipif(is_before_spark_320(), reason="Delta Lake writes are not supported before Spark 3.2.x")
@pytest.mark.xfail(condition=is_spark_340_or_later() and is_databricks_runtime(), reason="https://github.com/NVIDIA/spark-rapids/issues/9675")
def test_delta_write_encryption_hadoopconfig_fallback(spark_tmp_path, write_options):
data_path = spark_tmp_path + "/DELTA_DATA"
def setup_hadoop_confs(spark):
Expand All @@ -486,6 +491,7 @@ def reset_hadoop_confs(spark):
@ignore_order
@pytest.mark.parametrize('codec', ['gzip'])
@pytest.mark.skipif(is_before_spark_320(), reason="Delta Lake writes are not supported before Spark 3.2.x")
@pytest.mark.xfail(condition=is_spark_340_or_later() and is_databricks_runtime(), reason="https://github.com/NVIDIA/spark-rapids/issues/9675")
def test_delta_write_compression_fallback(spark_tmp_path, codec):
data_path = spark_tmp_path + "/DELTA_DATA"
confs=copy_and_update(delta_writes_enabled_conf, {"spark.sql.parquet.compression.codec": codec})
Expand All @@ -500,6 +506,7 @@ def test_delta_write_compression_fallback(spark_tmp_path, codec):
@delta_lake
@ignore_order
@pytest.mark.skipif(is_before_spark_320(), reason="Delta Lake writes are not supported before Spark 3.2.x")
@pytest.mark.xfail(condition=is_spark_340_or_later() and is_databricks_runtime(), reason="https://github.com/NVIDIA/spark-rapids/issues/9675")
def test_delta_write_legacy_format_fallback(spark_tmp_path):
data_path = spark_tmp_path + "/DELTA_DATA"
confs=copy_and_update(delta_writes_enabled_conf, {"spark.sql.parquet.writeLegacyFormat": "true"})
Expand Down Expand Up @@ -880,6 +887,7 @@ def test_delta_write_optimized_supported_types_partitioned(spark_tmp_path):
simple_string_to_string_map_gen,
StructGen([("x", ArrayGen(int_gen))]),
ArrayGen(StructGen([("x", long_gen)]))], ids=idfn)
@pytest.mark.xfail(condition=is_spark_340_or_later() and is_databricks_runtime(), reason="https://github.com/NVIDIA/spark-rapids/issues/9675")
def test_delta_write_optimized_unsupported_sort_fallback(spark_tmp_path, gen):
data_path = spark_tmp_path + "/DELTA_DATA"
confs=copy_and_update(delta_writes_enabled_conf, {
Expand Down

0 comments on commit 00b498e

Please sign in to comment.