lanyuanxiaoyao/hudi
1
0
Fork 0
Code Activity

[HUDI-1104] Adding support for UserDefinedPartitioners and SortModes to BulkInsert with Rows (#3149)

Browse Source
This commit is contained in:
Sivabalan Narayanan
2021-07-07 11:15:25 -04:00
committed by GitHub
parent 55ecbc662e
commit ea9e5d0e8b
31 changed files with 618 additions and 82 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/hudi/HoodieSparkSqlWriter.scala
View File

@@ -19,7 +19,6 @@ package org.apache.hudi
import java.util
import java.util.Properties
import org.apache.avro.generic.GenericRecord
import org.apache.hadoop.conf.Configuration
import org.apache.hadoop.fs.{FileSystem, Path}
@@ -34,13 +33,15 @@ import org.apache.hudi.common.table.{HoodieTableConfig, HoodieTableMetaClient}
import org.apache.hudi.common.table.timeline.HoodieActiveTimeline
import org.apache.hudi.common.util.{CommitUtils, ReflectionUtils}
import org.apache.hudi.config.HoodieBootstrapConfig.{BOOTSTRAP_BASE_PATH_PROP, BOOTSTRAP_INDEX_CLASS_PROP}
import org.apache.hudi.config.HoodieWriteConfig
import org.apache.hudi.config.{HoodieInternalConfig, HoodieWriteConfig}
import org.apache.hudi.exception.HoodieException
import org.apache.hudi.execution.bulkinsert.BulkInsertInternalPartitionerWithRowsFactory
import org.apache.hudi.hive.util.ConfigUtils
import org.apache.hudi.hive.{HiveSyncConfig, HiveSyncTool}
import org.apache.hudi.internal.DataSourceInternalWriterHelper
import org.apache.hudi.keygen.factory.HoodieSparkKeyGeneratorFactory
import org.apache.hudi.sync.common.AbstractSyncTool
import org.apache.hudi.table.BulkInsertPartitioner
import org.apache.log4j.LogManager
import org.apache.spark.SPARK_VERSION
import org.apache.spark.SparkContext
@@ -50,7 +51,7 @@ import org.apache.spark.sql.hudi.HoodieSqlUtils
import org.apache.spark.sql.internal.SQLConf
import org.apache.spark.sql.internal.StaticSQLConf.SCHEMA_STRING_LENGTH_THRESHOLD
import org.apache.spark.sql.types.StructType
import org.apache.spark.sql.{DataFrame, SQLContext, SaveMode, SparkSession}
import org.apache.spark.sql.{DataFrame, Dataset, Row, SQLContext, SaveMode, SparkSession}
import scala.collection.JavaConversions._
import scala.collection.mutable.ListBuffer
@@ -335,7 +336,17 @@ object HoodieSparkSqlWriter {
}
val params = parameters.updated(HoodieWriteConfig.AVRO_SCHEMA.key, schema.toString)
val writeConfig = DataSourceUtils.createHoodieConfig(schema.toString, path.get, tblName, mapAsJavaMap(params))
val hoodieDF = HoodieDatasetBulkInsertHelper.prepareHoodieDatasetForBulkInsert(sqlContext, writeConfig, df, structName, nameSpace)
val userDefinedBulkInsertPartitionerOpt = DataSourceUtils.createUserDefinedBulkInsertPartitionerWithRows(writeConfig)
val bulkInsertPartitionerRows : BulkInsertPartitioner[Dataset[Row]] = if (userDefinedBulkInsertPartitionerOpt.isPresent) {
userDefinedBulkInsertPartitionerOpt.get
}
else {
BulkInsertInternalPartitionerWithRowsFactory.get(writeConfig.getBulkInsertSortMode)
}
val arePartitionRecordsSorted = bulkInsertPartitionerRows.arePartitionRecordsSorted();
parameters.updated(HoodieInternalConfig.BULKINSERT_ARE_PARTITIONER_RECORDS_SORTED, arePartitionRecordsSorted.toString)
val hoodieDF = HoodieDatasetBulkInsertHelper.prepareHoodieDatasetForBulkInsert(sqlContext, writeConfig, df, structName, nameSpace,
bulkInsertPartitionerRows)
if (SPARK_VERSION.startsWith("2.")) {
hoodieDF.write.format("org.apache.hudi.internal")
.option(DataSourceInternalWriterHelper.INSTANT_TIME_OPT_KEY, instantTime)