Revert "[GLUTEN-8327][CORE] Introduce the ConfigEntry to make the con…

…fig definition more flexible (apache#8328)" This reverts commit 3c53f9f.
baibaichen · Dec 31, 2024 · 6800f39 · 6800f39
1 parent 7503ce8
commit 6800f39
Show file tree

Hide file tree

Showing 222 changed files with 633 additions and 1,138 deletions.
diff --git a/.../src-celeborn/main/scala/org/apache/spark/shuffle/CHCelebornColumnarBatchSerializer.scala b/.../src-celeborn/main/scala/org/apache/spark/shuffle/CHCelebornColumnarBatchSerializer.scala
@@ -16,8 +16,8 @@
  */
 package org.apache.spark.shuffle
 
+import org.apache.gluten.GlutenConfig
 import org.apache.gluten.backendsapi.clickhouse.CHBackendSettings
-import org.apache.gluten.config.GlutenConfig
 import org.apache.gluten.vectorized.BlockOutputStream
 import org.apache.gluten.vectorized.CHStreamReader
 
@@ -59,14 +59,14 @@ private class CHCelebornColumnarBatchSerializerInstance(
   with Logging {
 
   private lazy val conf = SparkEnv.get.conf
-  private lazy val gluten_conf = GlutenConfig.get
+  private lazy val gluten_conf = GlutenConfig.getConf
   private lazy val compressionCodec = GlutenShuffleUtils.getCompressionCodec(conf)
   private lazy val capitalizedCompressionCodec = compressionCodec.toUpperCase(Locale.ROOT)
   private lazy val compressionLevel =
     GlutenShuffleUtils.getCompressionLevel(
       conf,
       compressionCodec,
-      GlutenConfig.get.columnarShuffleCodecBackend.orNull)
+      GlutenConfig.getConf.columnarShuffleCodecBackend.orNull)
 
   override def deserializeStream(in: InputStream): DeserializationStream = {
     new DeserializationStream {

diff --git a/...se/src-celeborn/main/scala/org/apache/spark/shuffle/CHCelebornColumnarShuffleWriter.scala b/...se/src-celeborn/main/scala/org/apache/spark/shuffle/CHCelebornColumnarShuffleWriter.scala
@@ -16,8 +16,8 @@
  */
 package org.apache.spark.shuffle
 
+import org.apache.gluten.GlutenConfig
 import org.apache.gluten.backendsapi.clickhouse.CHBackendSettings
-import org.apache.gluten.config.GlutenConfig
 import org.apache.gluten.execution.ColumnarNativeIterator
 import org.apache.gluten.memory.CHThreadGroup
 import org.apache.gluten.vectorized._
@@ -80,10 +80,10 @@ class CHCelebornColumnarShuffleWriter[K, V](
       nativeBufferSize,
       capitalizedCompressionCodec,
       compressionLevel,
-      GlutenConfig.get.chColumnarShuffleSpillThreshold,
+      GlutenConfig.getConf.chColumnarShuffleSpillThreshold,
       CHBackendSettings.shuffleHashAlgorithm,
       celebornPartitionPusher,
-      GlutenConfig.get.chColumnarForceMemorySortShuffle
+      GlutenConfig.getConf.chColumnarForceMemorySortShuffle
         || ShuffleMode.SORT.name.equalsIgnoreCase(shuffleWriterType)
     )
 

diff --git a/.../src-delta-32/main/scala/org/apache/spark/sql/delta/ClickhouseOptimisticTransaction.scala b/.../src-delta-32/main/scala/org/apache/spark/sql/delta/ClickhouseOptimisticTransaction.scala
@@ -16,8 +16,8 @@
  */
 package org.apache.spark.sql.delta
 
+import org.apache.gluten.GlutenConfig
 import org.apache.gluten.backendsapi.clickhouse.CHConf
-import org.apache.gluten.config.GlutenConfig
 
 import org.apache.spark.SparkException
 import org.apache.spark.sql.Dataset
@@ -68,7 +68,7 @@ class ClickhouseOptimisticTransaction(
       writeOptions: Option[DeltaOptions],
       isOptimize: Boolean,
       additionalConstraints: Seq[Constraint]): Seq[FileAction] = {
-    val nativeWrite = GlutenConfig.get.enableNativeWriter.getOrElse(false)
+    val nativeWrite = GlutenConfig.getConf.enableNativeWriter.getOrElse(false)
     if (writingMergeTree) {
       // TODO: update FallbackByBackendSettings for mergetree always return true
       val onePipeline = nativeWrite && CHConf.get.enableOnePipelineMergeTreeWrite

diff --git a/...c-delta-32/test/scala/org/apache/spark/sql/execution/datasources/DeltaV1WritesSuite.scala b/...c-delta-32/test/scala/org/apache/spark/sql/execution/datasources/DeltaV1WritesSuite.scala
@@ -16,7 +16,7 @@
  */
 package org.apache.spark.sql.execution.datasources
 
-import org.apache.gluten.config.GlutenConfig
+import org.apache.gluten.GlutenConfig
 import org.apache.gluten.execution.{GlutenClickHouseWholeStageTransformerSuite, GlutenPlan, SortExecTransformer}
 import org.apache.spark.SparkConf
 import org.apache.spark.sql.execution.datasources.parquet.ParquetFileFormat

diff --git a/...e/src-iceberg/test/scala/org/apache/gluten/execution/iceberg/ClickHouseIcebergSuite.scala b/...e/src-iceberg/test/scala/org/apache/gluten/execution/iceberg/ClickHouseIcebergSuite.scala
@@ -17,7 +17,7 @@
 package org.apache.gluten.execution.iceberg
 
 import org.apache.gluten.execution.{GlutenClickHouseWholeStageTransformerSuite, IcebergScanTransformer}
-import org.apache.gluten.config.GlutenConfig
+import org.apache.gluten.GlutenConfig
 import org.apache.spark.SparkConf
 import org.apache.spark.sql.Row
 
@@ -62,7 +62,7 @@ class ClickHouseIcebergSuite extends GlutenClickHouseWholeStageTransformerSuite
     val rightTable = "p_int_tb"
     withTable(leftTable, rightTable) {
       // Partition key of string type.
-      withSQLConf(GlutenConfig.GLUTEN_ENABLED.key -> "false") {
+      withSQLConf(GlutenConfig.GLUTEN_ENABLED_KEY -> "false") {
         // Gluten does not support write iceberg table.
         spark.sql(
           s"""
@@ -84,7 +84,7 @@ class ClickHouseIcebergSuite extends GlutenClickHouseWholeStageTransformerSuite
 
       // Partition key of integer type.
       withSQLConf(
-        GlutenConfig.GLUTEN_ENABLED.key -> "false"
+        GlutenConfig.GLUTEN_ENABLED_KEY -> "false"
       ) {
         // Gluten does not support write iceberg table.
         spark.sql(
@@ -145,7 +145,7 @@ class ClickHouseIcebergSuite extends GlutenClickHouseWholeStageTransformerSuite
     val rightTable = "p_int_tb"
     withTable(leftTable, rightTable) {
       // Partition key of string type.
-      withSQLConf(GlutenConfig.GLUTEN_ENABLED.key -> "false") {
+      withSQLConf(GlutenConfig.GLUTEN_ENABLED_KEY -> "false") {
         // Gluten does not support write iceberg table.
         spark.sql(
           s"""
@@ -167,7 +167,7 @@ class ClickHouseIcebergSuite extends GlutenClickHouseWholeStageTransformerSuite
 
       // Partition key of integer type.
       withSQLConf(
-        GlutenConfig.GLUTEN_ENABLED.key -> "false"
+        GlutenConfig.GLUTEN_ENABLED_KEY -> "false"
       ) {
         // Gluten does not support write iceberg table.
         spark.sql(
@@ -228,7 +228,7 @@ class ClickHouseIcebergSuite extends GlutenClickHouseWholeStageTransformerSuite
     val rightTable = "p_int_tb"
     withTable(leftTable, rightTable) {
       // Partition key of string type.
-      withSQLConf(GlutenConfig.GLUTEN_ENABLED.key -> "false") {
+      withSQLConf(GlutenConfig.GLUTEN_ENABLED_KEY -> "false") {
         // Gluten does not support write iceberg table.
         spark.sql(
           s"""
@@ -250,7 +250,7 @@ class ClickHouseIcebergSuite extends GlutenClickHouseWholeStageTransformerSuite
 
       // Partition key of integer type.
       withSQLConf(
-        GlutenConfig.GLUTEN_ENABLED.key -> "false"
+        GlutenConfig.GLUTEN_ENABLED_KEY -> "false"
       ) {
         // Gluten does not support write iceberg table.
         spark.sql(
@@ -350,7 +350,7 @@ class ClickHouseIcebergSuite extends GlutenClickHouseWholeStageTransformerSuite
 
   test("iceberg read mor table - delete and update") {
     withTable("iceberg_mor_tb") {
-      withSQLConf(GlutenConfig.GLUTEN_ENABLED.key -> "false") {
+      withSQLConf(GlutenConfig.GLUTEN_ENABLED_KEY -> "false") {
         spark.sql(
           """
             |create table iceberg_mor_tb (
@@ -403,7 +403,7 @@ class ClickHouseIcebergSuite extends GlutenClickHouseWholeStageTransformerSuite
   // TODO: support merge-on-read mode
   ignore("iceberg read mor table - delete and update with merge-on-read mode") {
     withTable("iceberg_mor_tb") {
-      withSQLConf(GlutenConfig.GLUTEN_ENABLED.key -> "false") {
+      withSQLConf(GlutenConfig.GLUTEN_ENABLED_KEY -> "false") {
         spark.sql(
           """
             |create table iceberg_mor_tb (
@@ -458,7 +458,7 @@ class ClickHouseIcebergSuite extends GlutenClickHouseWholeStageTransformerSuite
 
   test("iceberg read mor table - merge into") {
     withTable("iceberg_mor_tb", "merge_into_source_tb") {
-      withSQLConf(GlutenConfig.GLUTEN_ENABLED.key -> "false") {
+      withSQLConf(GlutenConfig.GLUTEN_ENABLED_KEY -> "false") {
         spark.sql(
           """
             |create table iceberg_mor_tb (
@@ -531,7 +531,7 @@ class ClickHouseIcebergSuite extends GlutenClickHouseWholeStageTransformerSuite
   // TODO: support merge-on-read mode
   ignore("iceberg read mor table - merge into with merge-on-read mode") {
     withTable("iceberg_mor_tb", "merge_into_source_tb") {
-      withSQLConf(GlutenConfig.GLUTEN_ENABLED.key -> "false") {
+      withSQLConf(GlutenConfig.GLUTEN_ENABLED_KEY -> "false") {
         spark.sql(
           """
             |create table iceberg_mor_tb (

diff --git a/...ds-clickhouse/src/main/java/org/apache/gluten/vectorized/CHNativeExpressionEvaluator.java b/...ds-clickhouse/src/main/java/org/apache/gluten/vectorized/CHNativeExpressionEvaluator.java
@@ -16,8 +16,8 @@
  */
 package org.apache.gluten.vectorized;
 
+import org.apache.gluten.GlutenConfig;
 import org.apache.gluten.backendsapi.BackendsApiManager;
-import org.apache.gluten.config.GlutenConfig;
 import org.apache.gluten.execution.ColumnarNativeIterator;
 import org.apache.gluten.memory.CHThreadGroup;
 import org.apache.gluten.utils.ConfigUtil;

diff --git a/backends-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHBackend.scala b/backends-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHBackend.scala
@@ -17,10 +17,10 @@
 package org.apache.gluten.backendsapi.clickhouse
 
 import org.apache.gluten.GlutenBuildInfo._
+import org.apache.gluten.GlutenConfig
 import org.apache.gluten.backendsapi._
 import org.apache.gluten.columnarbatch.CHBatch
 import org.apache.gluten.component.Component.BuildInfo
-import org.apache.gluten.config.GlutenConfig
 import org.apache.gluten.execution.WriteFilesExecTransformer
 import org.apache.gluten.expression.WindowFunctionsBuilder
 import org.apache.gluten.extension.ValidationResult
@@ -282,11 +282,11 @@ object CHBackendSettings extends BackendSettingsApi with Logging {
   }
 
   override def supportSortExec(): Boolean = {
-    GlutenConfig.get.enableColumnarSort
+    GlutenConfig.getConf.enableColumnarSort
   }
 
   override def supportSortMergeJoinExec(): Boolean = {
-    GlutenConfig.get.enableColumnarSortMergeJoin
+    GlutenConfig.getConf.enableColumnarSortMergeJoin
   }
 
   override def supportWindowExec(windowFunctions: Seq[NamedExpression]): Boolean = {
@@ -391,7 +391,7 @@ object CHBackendSettings extends BackendSettingsApi with Logging {
   }
 
   override def enableNativeWriteFiles(): Boolean = {
-    GlutenConfig.get.enableNativeWriter.getOrElse(false)
+    GlutenConfig.getConf.enableNativeWriter.getOrElse(false)
   }
 
   override def supportCartesianProductExec(): Boolean = true

diff --git a/backends-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHConf.scala b/backends-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHConf.scala
@@ -16,7 +16,7 @@
  */
 package org.apache.gluten.backendsapi.clickhouse
 
-import org.apache.gluten.config.GlutenConfig
+import org.apache.gluten.GlutenConfig
 
 import org.apache.spark.SparkConf
 import org.apache.spark.sql.internal.SQLConf

diff --git a/...ds-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHIteratorApi.scala b/...ds-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHIteratorApi.scala
@@ -16,8 +16,8 @@
  */
 package org.apache.gluten.backendsapi.clickhouse
 
+import org.apache.gluten.GlutenNumaBindingInfo
 import org.apache.gluten.backendsapi.IteratorApi
-import org.apache.gluten.config.GlutenNumaBindingInfo
 import org.apache.gluten.execution._
 import org.apache.gluten.expression.ConverterUtils
 import org.apache.gluten.logging.LogLevelUtil

diff --git a/...ds-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHListenerApi.scala b/...ds-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHListenerApi.scala
@@ -16,9 +16,9 @@
  */
 package org.apache.gluten.backendsapi.clickhouse
 
+import org.apache.gluten.GlutenConfig
 import org.apache.gluten.backendsapi.ListenerApi
 import org.apache.gluten.columnarbatch.CHBatch
-import org.apache.gluten.config.GlutenConfig
 import org.apache.gluten.execution.CHBroadcastBuildSideCache
 import org.apache.gluten.execution.datasource.GlutenFormatFactory
 import org.apache.gluten.expression.UDFMappings

diff --git a/backends-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHRuleApi.scala b/backends-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHRuleApi.scala
@@ -16,9 +16,9 @@
  */
 package org.apache.gluten.backendsapi.clickhouse
 
+import org.apache.gluten.GlutenConfig
 import org.apache.gluten.backendsapi.RuleApi
 import org.apache.gluten.columnarbatch.CHBatch
-import org.apache.gluten.config.GlutenConfig
 import org.apache.gluten.extension._
 import org.apache.gluten.extension.columnar._
 import org.apache.gluten.extension.columnar.MiscColumnarRules.{RemoveGlutenTableCacheColumnarToRow, RemoveTopmostColumnarToRow, RewriteSubqueryBroadcast}

diff --git a/...ickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHSparkPlanExecApi.scala b/...ickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHSparkPlanExecApi.scala
@@ -16,8 +16,8 @@
  */
 package org.apache.gluten.backendsapi.clickhouse
 
+import org.apache.gluten.GlutenConfig
 import org.apache.gluten.backendsapi.{BackendsApiManager, SparkPlanExecApi}
-import org.apache.gluten.config.GlutenConfig
 import org.apache.gluten.exception.{GlutenException, GlutenNotSupportException}
 import org.apache.gluten.execution._
 import org.apache.gluten.expression._
@@ -452,12 +452,12 @@ class CHSparkPlanExecApi extends SparkPlanExecApi with Logging {
     val readBatchNumRows = metrics("avgReadBatchNumRows")
     val numOutputRows = metrics("numOutputRows")
     val dataSize = metrics("dataSize")
-    if (GlutenConfig.get.isUseCelebornShuffleManager) {
+    if (GlutenConfig.getConf.isUseCelebornShuffleManager) {
       val clazz = ClassUtils.getClass("org.apache.spark.shuffle.CHCelebornColumnarBatchSerializer")
       val constructor =
         clazz.getConstructor(classOf[SQLMetric], classOf[SQLMetric], classOf[SQLMetric])
       constructor.newInstance(readBatchNumRows, numOutputRows, dataSize).asInstanceOf[Serializer]
-    } else if (GlutenConfig.get.isUseUniffleShuffleManager) {
+    } else if (GlutenConfig.getConf.isUseUniffleShuffleManager) {
       throw new UnsupportedOperationException("temporarily uniffle not support ch ")
     } else {
       new CHColumnarBatchSerializer(readBatchNumRows, numOutputRows, dataSize)

diff --git a/...s-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHValidatorApi.scala b/...s-clickhouse/src/main/scala/org/apache/gluten/backendsapi/clickhouse/CHValidatorApi.scala
@@ -16,8 +16,8 @@
  */
 package org.apache.gluten.backendsapi.clickhouse
 
+import org.apache.gluten.GlutenConfig
 import org.apache.gluten.backendsapi.ValidatorApi
-import org.apache.gluten.config.GlutenConfig
 import org.apache.gluten.expression.ExpressionConverter
 import org.apache.gluten.extension.ValidationResult
 import org.apache.gluten.substrait.SubstraitContext
@@ -95,7 +95,7 @@ class CHValidatorApi extends ValidatorApi with AdaptiveSparkPlanHelper with Logg
         }
       case rangePartitoning: RangePartitioning =>
         if (
-          GlutenConfig.get.enableColumnarSort &&
+          GlutenConfig.getConf.enableColumnarSort &&
           RangePartitionerBoundsGenerator.supportedOrderings(rangePartitoning, child)
         ) {
           None

diff --git a/...ckhouse/src/main/scala/org/apache/gluten/extension/CommonSubexpressionEliminateRule.scala b/...ckhouse/src/main/scala/org/apache/gluten/extension/CommonSubexpressionEliminateRule.scala
@@ -16,7 +16,7 @@
  */
 package org.apache.gluten.extension
 
-import org.apache.gluten.config.GlutenConfig
+import org.apache.gluten.GlutenConfig
 
 import org.apache.spark.internal.Logging
 import org.apache.spark.sql.SparkSession
@@ -39,8 +39,8 @@ class CommonSubexpressionEliminateRule(spark: SparkSession) extends Rule[Logical
   override def apply(plan: LogicalPlan): LogicalPlan = {
     val newPlan =
       if (
-        plan.resolved && GlutenConfig.get.enableGluten
-        && GlutenConfig.get.enableCommonSubexpressionEliminate && !plan.fastEquals(lastPlan)
+        plan.resolved && GlutenConfig.getConf.enableGluten
+        && GlutenConfig.getConf.enableCommonSubexpressionEliminate && !plan.fastEquals(lastPlan)
       ) {
         lastPlan = plan
         visitPlan(plan)

diff --git a/...ds-clickhouse/src/main/scala/org/apache/gluten/extension/CountDistinctWithoutExpand.scala b/...ds-clickhouse/src/main/scala/org/apache/gluten/extension/CountDistinctWithoutExpand.scala
@@ -16,7 +16,7 @@
  */
 package org.apache.gluten.extension
 
-import org.apache.gluten.config.GlutenConfig
+import org.apache.gluten.GlutenConfig
 
 import org.apache.spark.sql.catalyst.expressions.aggregate.{AggregateExpression, Count, CountDistinct}
 import org.apache.spark.sql.catalyst.plans.logical.LogicalPlan
@@ -32,7 +32,9 @@ import org.apache.spark.sql.catalyst.trees.TreePattern.AGGREGATE_EXPRESSION
  */
 object CountDistinctWithoutExpand extends Rule[LogicalPlan] {
   def apply(plan: LogicalPlan): LogicalPlan = {
-    if (GlutenConfig.get.enableGluten && GlutenConfig.get.enableCountDistinctWithoutExpand) {
+    if (
+      GlutenConfig.getConf.enableGluten && GlutenConfig.getConf.enableCountDistinctWithoutExpand
+    ) {
       plan.transformAllExpressionsWithPruning(_.containsPattern(AGGREGATE_EXPRESSION)) {
         case ae: AggregateExpression
             if ae.isDistinct && ae.aggregateFunction.isInstanceOf[Count] &&

diff --git a/backends-clickhouse/src/main/scala/org/apache/gluten/extension/ExtendedColumnPruning.scala b/backends-clickhouse/src/main/scala/org/apache/gluten/extension/ExtendedColumnPruning.scala
@@ -16,7 +16,7 @@
  */
 package org.apache.gluten.extension
 
-import org.apache.gluten.config.GlutenConfig
+import org.apache.gluten.GlutenConfig
 
 import org.apache.spark.internal.Logging
 import org.apache.spark.sql.SparkSession
@@ -36,7 +36,7 @@ object ExtendedGeneratorNestedColumnAliasing {
   def unapply(plan: LogicalPlan): Option[LogicalPlan] = plan match {
     case pj @ Project(projectList, f @ Filter(condition, g: Generate))
         if canPruneGenerator(g.generator) &&
-          GlutenConfig.get.enableExtendedColumnPruning &&
+          GlutenConfig.getConf.enableExtendedColumnPruning &&
           (SQLConf.get.nestedPruningOnExpressions || SQLConf.get.nestedSchemaPruningEnabled) =>
       val attrToExtractValues =
         getAttributeToExtractValues(projectList ++ g.generator.children :+ condition, Seq.empty)