ajantha-bhat commented on a change in pull request #3787:
URL: https://github.com/apache/carbondata/pull/3787#discussion_r486218886
##########
File path:
integration/spark/src/main/scala/org/apache/spark/sql/secondaryindex/rdd/SecondaryIndexCreator.scala
##########
@@ -428,4 +552,40 @@ object SecondaryIndexCreator {
}
threadPoolSize
}
+
+ def dataFrameOfSegments(
+ sparkSession: SparkSession,
+ carbonTable: CarbonTable,
+ projections: String,
+ segments: Array[String]): DataFrame = {
+ try {
+ CarbonUtils
+ .threadSet(CarbonCommonConstants.CARBON_INPUT_SEGMENTS +
+ carbonTable.getDatabaseName + CarbonCommonConstants.POINT +
+ carbonTable.getTableName,
+ segments.mkString(","))
Review comment:
Moved. These are created by reformat command itself (ctrl + alt + shift
+ L), so need the correct tool to properly reformat or not use it.
##########
File path:
integration/spark/src/main/scala/org/apache/spark/sql/secondaryindex/rdd/SecondaryIndexCreator.scala
##########
@@ -152,68 +158,181 @@ object SecondaryIndexCreator {
LOGGER.info("spark.dynamicAllocation.maxExecutors property is set to
=" + execInstance)
}
}
- var futureObjectList = List[java.util.concurrent.Future[Array[(String,
Boolean)]]]()
- for (eachSegment <- validSegmentList) {
- val segId = eachSegment
- futureObjectList :+= executorService.submit(new
Callable[Array[(String, Boolean)]] {
- @throws(classOf[Exception])
- override def call(): Array[(String, Boolean)] = {
-
ThreadLocalSessionInfo.getOrCreateCarbonSessionInfo().getNonSerializableExtraInfo
- .put("carbonConf",
SparkSQLUtil.sessionState(sc.sparkSession).newHadoopConf())
- var eachSegmentSecondaryIndexCreationStatus: Array[(String,
Boolean)] = Array.empty
- CarbonLoaderUtil.checkAndCreateCarbonDataLocation(segId,
indexCarbonTable)
- val carbonLoadModel = getCopyObject(secondaryIndexModel)
- carbonLoadModel
-
.setFactTimeStamp(secondaryIndexModel.segmentIdToLoadStartTimeMapping(eachSegment))
-
carbonLoadModel.setTablePath(secondaryIndexModel.carbonTable.getTablePath)
- val secondaryIndexCreationStatus = new
CarbonSecondaryIndexRDD(sc.sparkSession,
- new SecondaryIndexCreationResultImpl,
- carbonLoadModel,
- secondaryIndexModel.secondaryIndex,
- segId, execInstance, indexCarbonTable, forceAccessSegment,
isCompactionCall).collect()
+ var successSISegments: List[String] = List()
+ var failedSISegments: List[String] = List()
+ val sort_scope =
indexCarbonTable.getTableInfo.getFactTable.getTableProperties
+ .get("sort_scope")
+ if (sort_scope != null && sort_scope.equalsIgnoreCase("global_sort")) {
+ val mainTable =
secondaryIndexModel.carbonLoadModel.getCarbonDataLoadSchema.getCarbonTable
+ var futureObjectList = List[java.util.concurrent.Future[Array[(String,
+ (LoadMetadataDetails, ExecutionErrors))]]]()
+ for (eachSegment <- validSegmentList) {
+ futureObjectList :+= executorService
+ .submit(new Callable[Array[(String, (LoadMetadataDetails,
ExecutionErrors))]] {
+ @throws(classOf[Exception])
+ override def call(): Array[(String, (LoadMetadataDetails,
ExecutionErrors))] = {
+ val carbonLoadModel = getCopyObject(secondaryIndexModel)
+ // loading, we need to query main table add position reference
+ val proj = indexCarbonTable.getCreateOrderColumn
+ .asScala
+ .map(_.getColName)
+ .filterNot(_.equals("positionReference")).toSet
Review comment:
done
##########
File path:
integration/spark/src/main/scala/org/apache/spark/sql/secondaryindex/rdd/SecondaryIndexCreator.scala
##########
@@ -152,68 +158,181 @@ object SecondaryIndexCreator {
LOGGER.info("spark.dynamicAllocation.maxExecutors property is set to
=" + execInstance)
}
}
- var futureObjectList = List[java.util.concurrent.Future[Array[(String,
Boolean)]]]()
- for (eachSegment <- validSegmentList) {
- val segId = eachSegment
- futureObjectList :+= executorService.submit(new
Callable[Array[(String, Boolean)]] {
- @throws(classOf[Exception])
- override def call(): Array[(String, Boolean)] = {
-
ThreadLocalSessionInfo.getOrCreateCarbonSessionInfo().getNonSerializableExtraInfo
- .put("carbonConf",
SparkSQLUtil.sessionState(sc.sparkSession).newHadoopConf())
- var eachSegmentSecondaryIndexCreationStatus: Array[(String,
Boolean)] = Array.empty
- CarbonLoaderUtil.checkAndCreateCarbonDataLocation(segId,
indexCarbonTable)
- val carbonLoadModel = getCopyObject(secondaryIndexModel)
- carbonLoadModel
-
.setFactTimeStamp(secondaryIndexModel.segmentIdToLoadStartTimeMapping(eachSegment))
-
carbonLoadModel.setTablePath(secondaryIndexModel.carbonTable.getTablePath)
- val secondaryIndexCreationStatus = new
CarbonSecondaryIndexRDD(sc.sparkSession,
- new SecondaryIndexCreationResultImpl,
- carbonLoadModel,
- secondaryIndexModel.secondaryIndex,
- segId, execInstance, indexCarbonTable, forceAccessSegment,
isCompactionCall).collect()
+ var successSISegments: List[String] = List()
+ var failedSISegments: List[String] = List()
+ val sort_scope =
indexCarbonTable.getTableInfo.getFactTable.getTableProperties
+ .get("sort_scope")
+ if (sort_scope != null && sort_scope.equalsIgnoreCase("global_sort")) {
+ val mainTable =
secondaryIndexModel.carbonLoadModel.getCarbonDataLoadSchema.getCarbonTable
+ var futureObjectList = List[java.util.concurrent.Future[Array[(String,
+ (LoadMetadataDetails, ExecutionErrors))]]]()
+ for (eachSegment <- validSegmentList) {
+ futureObjectList :+= executorService
+ .submit(new Callable[Array[(String, (LoadMetadataDetails,
ExecutionErrors))]] {
+ @throws(classOf[Exception])
+ override def call(): Array[(String, (LoadMetadataDetails,
ExecutionErrors))] = {
+ val carbonLoadModel = getCopyObject(secondaryIndexModel)
+ // loading, we need to query main table add position reference
Review comment:
done
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
