Github user jackylk commented on a diff in the pull request:
https://github.com/apache/incubator-carbondata/pull/203#discussion_r80915025
--- Diff:
integration/spark/src/main/scala/org/apache/carbondata/spark/rdd/CarbonDataLoadRDD.scala
---
@@ -145,92 +232,28 @@ class CarbonDataLoadRDD[K, V](
override def compute(theSplit: Partition, context: TaskContext):
Iterator[(K, V)] = {
val LOGGER = LogServiceFactory.getLogService(this.getClass.getName)
val iter = new Iterator[(K, V)] {
- var dataloadStatus = CarbonCommonConstants.STORE_LOADSTATUS_FAILURE
var partitionID = "0"
+ val loadMetadataDetails = new LoadMetadataDetails()
var model: CarbonLoadModel = _
var uniqueLoadStatusId = carbonLoadModel.getTableName +
CarbonCommonConstants.UNDERSCORE +
theSplit.index
try {
- val carbonPropertiesFilePath =
System.getProperty("carbon.properties.filepath", null)
- if (null == carbonPropertiesFilePath) {
- System.setProperty("carbon.properties.filepath",
- System.getProperty("user.dir") + '/' + "conf" + '/' +
"carbon.properties")
- }
+ loadMetadataDetails.setPartitionCount(partitionID)
+
loadMetadataDetails.setLoadStatus(CarbonCommonConstants.STORE_LOADSTATUS_FAILURE)
+
carbonLoadModel.setSegmentId(String.valueOf(loadCount))
setModelAndBlocksInfo()
-
CarbonTimeStatisticsFactory.getLoadStatisticsInstance.initPartitonInfo(model.getPartitionId)
-
CarbonProperties.getInstance().addProperty("carbon.is.columnar.storage", "true")
-
CarbonProperties.getInstance().addProperty("carbon.dimension.split.value.in.columnar",
"1")
-
CarbonProperties.getInstance().addProperty("carbon.is.fullyfilled.bits", "true")
- CarbonProperties.getInstance().addProperty("is.int.based.indexer",
"true")
-
CarbonProperties.getInstance().addProperty("aggregate.columnar.keyblock",
"true")
-
CarbonProperties.getInstance().addProperty("high.cardinality.value", "100000")
-
CarbonProperties.getInstance().addProperty("is.compressed.keyblock", "false")
-
CarbonProperties.getInstance().addProperty("carbon.leaf.node.size", "120000")
-
- // this property is used to determine whether temp location for
carbon is inside
- // container temp dir or is yarn application directory.
- val carbonUseLocalDir = CarbonProperties.getInstance()
- .getProperty("carbon.use.local.dir", "false")
-
- if(carbonUseLocalDir.equalsIgnoreCase("true")) {
- val storeLocations =
CarbonLoaderUtil.getConfiguredLocalDirs(SparkEnv.get.conf)
- if (null != storeLocations && storeLocations.nonEmpty) {
- storeLocation =
storeLocations(Random.nextInt(storeLocations.length))
- }
- if (storeLocation == null) {
- storeLocation = System.getProperty("java.io.tmpdir")
- }
- }
- else {
- storeLocation = System.getProperty("java.io.tmpdir")
- }
- storeLocation = storeLocation + '/' + System.nanoTime() + '/' +
theSplit.index
- dataloadStatus = CarbonCommonConstants.STORE_LOADSTATUS_SUCCESS
-
+ storeLocation = CarbonDataLoadRDD.initialize(model, theSplit.index)
+
loadMetadataDetails.setLoadStatus(CarbonCommonConstants.STORE_LOADSTATUS_SUCCESS)
if (model.isRetentionRequest) {
recreateAggregationTableForRetention
}
else if (model.isAggLoadRequest) {
--- End diff --
Is this required? If it is not required, remove it.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
