From e05ed1b893b43b6e6a69e892af23a817370829d7 Mon Sep 17 00:00:00 2001
From: Thomas Graves <tgraves@nvidia.com>
Date: Fri, 30 Jul 2021 11:13:09 -0500
Subject: [PATCH] Update spark 301db shim to new ParquetFilter api (#3100)

Signed-off-by: Thomas Graves <tgraves@nvidia.com>
---
 .../rapids/shims/spark301db/Spark301dbShims.scala | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/shims/spark301db/src/main/scala/com/nvidia/spark/rapids/shims/spark301db/Spark301dbShims.scala b/shims/spark301db/src/main/scala/com/nvidia/spark/rapids/shims/spark301db/Spark301dbShims.scala
index 984b22e3f11..856e4fd8b50 100644
--- a/shims/spark301db/src/main/scala/com/nvidia/spark/rapids/shims/spark301db/Spark301dbShims.scala
+++ b/shims/spark301db/src/main/scala/com/nvidia/spark/rapids/shims/spark301db/Spark301dbShims.scala
@@ -20,6 +20,7 @@ import com.databricks.sql.execution.window.RunningWindowFunctionExec
 import com.nvidia.spark.rapids._
 import com.nvidia.spark.rapids.shims.spark301.Spark301Shims
 import org.apache.hadoop.fs.Path
+import org.apache.parquet.schema.MessageType
 
 import org.apache.spark.sql.rapids.shims.spark301db._
 import org.apache.spark.rdd.RDD
@@ -33,11 +34,13 @@ import org.apache.spark.sql.execution._
 import org.apache.spark.sql.execution.adaptive.ShuffleQueryStageExec
 import org.apache.spark.sql.execution.datasources.{FilePartition, HadoopFsRelation, PartitionDirectory, PartitionedFile}
 import org.apache.spark.sql.execution.datasources.json.JsonFileFormat
+import org.apache.spark.sql.execution.datasources.parquet.ParquetFilters
 import org.apache.spark.sql.execution.exchange.ShuffleExchangeExec
 import org.apache.spark.sql.execution.joins.{BroadcastHashJoinExec, BroadcastNestedLoopJoinExec, HashJoin, SortMergeJoinExec}
 import org.apache.spark.sql.execution.joins.ShuffledHashJoinExec
 import org.apache.spark.sql.execution.python.{AggregateInPandasExec, ArrowEvalPythonExec, FlatMapGroupsInPandasExec, MapInPandasExec, WindowInPandasExec}
 import org.apache.spark.sql.execution.window.WindowExecBase
+import org.apache.spark.sql.internal.SQLConf
 import org.apache.spark.sql.rapids.GpuFileSourceScanExec
 import org.apache.spark.sql.rapids.execution.{GpuBroadcastExchangeExecBase, GpuBroadcastNestedLoopJoinExecBase, GpuShuffleExchangeExecBase}
 import org.apache.spark.sql.rapids.execution.python.{GpuAggregateInPandasExecMeta, GpuArrowEvalPythonExec, GpuFlatMapGroupsInPandasExecMeta, GpuMapInPandasExecMeta, GpuPythonUDF, GpuWindowInPandasExecMetaBase}
@@ -47,6 +50,18 @@ class Spark301dbShims extends Spark301Shims {
 
   override def getSparkShimVersion: ShimVersion = SparkShimServiceProvider.VERSION
 
+  override def getParquetFilters(
+      schema: MessageType,
+      pushDownDate: Boolean,
+      pushDownTimestamp: Boolean,
+      pushDownDecimal: Boolean,
+      pushDownStartWith: Boolean,
+      pushDownInFilterThreshold: Int,
+      caseSensitive: Boolean,
+      datetimeRebaseMode: SQLConf.LegacyBehaviorPolicy.Value): ParquetFilters =
+    new ParquetFilters(schema, pushDownDate, pushDownTimestamp, pushDownDecimal, pushDownStartWith,
+      pushDownInFilterThreshold, caseSensitive, datetimeRebaseMode)
+
   override def getGpuBroadcastNestedLoopJoinShim(
       left: SparkPlan,
       right: SparkPlan,