Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Avoid unapply on PromotePrecision [databricks] #4583

Merged
merged 2 commits into from
Jan 21, 2022
Merged

Avoid unapply on PromotePrecision [databricks] #4583

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
90584ba
Avoid unapply on PromotePrecision
jlowe Jan 20, 2022
b4ceb5a
Skip tests that hang on DB9.1
jlowe Jan 21, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions integration_tests/src/main/python/udf_test.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@
import pytest

from conftest import is_at_least_precommit_run
from spark_session import is_databricks91_or_later

from pyspark.sql.pandas.utils import require_minimum_pyarrow_version, require_minimum_pandas_version
try:
Expand Down Expand Up @@ -212,6 +213,7 @@ def pandas_sum(to_process: pd.Series) -> list:


# ======= Test flat map group in Pandas =======
@pytest.mark.skipif(is_databricks91_or_later(), reason="https://github.com/NVIDIA/spark-rapids/issues/4599")
@ignore_order
@pytest.mark.parametrize('data_gen', [LongGen()], ids=idfn)
def test_group_apply_udf(data_gen):
Expand All @@ -226,6 +228,7 @@ def pandas_add(data):
conf=arrow_udf_conf)


@pytest.mark.skipif(is_databricks91_or_later(), reason="https://github.com/NVIDIA/spark-rapids/issues/4599")
@ignore_order
@pytest.mark.parametrize('data_gen', arrow_common_gen, ids=idfn)
def test_group_apply_udf_more_types(data_gen):
Expand Down
22 changes: 4 additions & 18 deletions sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuOverrides.scala
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -492,21 +492,6 @@ object GpuOverrides extends Logging {
!regexList.exists(pattern => s.contains(pattern))
}

private def convertExprToGpuIfPossible(expr: Expression, conf: RapidsConf): Expression = {
if (expr.find(_.isInstanceOf[GpuExpression]).isDefined) {
// already been converted
expr
} else {
val wrapped = wrapExpr(expr, conf, None)
wrapped.tagForGpu()
if (wrapped.canExprTreeBeReplaced) {
wrapped.convertToGpu()
} else {
expr
}
}
}

private def convertPartToGpuIfPossible(part: Partitioning, conf: RapidsConf): Partitioning = {
part match {
case _: GpuPartitioning => part
Expand Down Expand Up @@ -900,7 +885,6 @@ object GpuOverrides extends Logging {
private[this] def extractOrigParam(expr: BaseExprMeta[_]): BaseExprMeta[_] =
expr.wrapped match {
case lit: Literal if lit.dataType.isInstanceOf[DecimalType] =>
val dt = lit.dataType.asInstanceOf[DecimalType]
// Lets figure out if we can make the Literal value smaller
val (newType, value) = lit.value match {
case null =>
Expand All @@ -920,8 +904,10 @@ object GpuOverrides extends Logging {
throw new IllegalArgumentException(s"Unexpected decimal literal value $other")
}
expr.asInstanceOf[LiteralExprMeta].withNewLiteral(Literal(value, newType))
// We avoid unapply for Cast because it changes between versions of Spark
case PromotePrecision(c: CastBase) if c.dataType.isInstanceOf[DecimalType] =>
// Avoid unapply for PromotePrecision and Cast because it changes between Spark versions
case p: PromotePrecision if p.child.isInstanceOf[CastBase] &&
p.child.dataType.isInstanceOf[DecimalType] =>
val c = p.child.asInstanceOf[CastBase]
val to = c.dataType.asInstanceOf[DecimalType]
val fromType = DecimalUtil.optionallyAsDecimalType(c.child.dataType)
fromType match {
Expand Down