garyli1019 commented on a change in pull request #1848: URL: https://github.com/apache/hudi/pull/1848#discussion_r464842847
########## File path: hudi-spark/src/main/scala/org/apache/hudi/HoodieMergeOnReadRDD.scala ########## @@ -0,0 +1,274 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.hudi + +import org.apache.hudi.common.fs.FSUtils +import org.apache.hudi.common.table.log.HoodieMergedLogRecordScanner +import org.apache.hudi.exception.HoodieException +import org.apache.hudi.hadoop.config.HoodieRealtimeConfig +import org.apache.hudi.hadoop.utils.HoodieInputFormatUtils.HOODIE_RECORD_KEY_COL_POS + +import org.apache.avro.Schema +import org.apache.avro.generic.{GenericRecord, GenericRecordBuilder} +import org.apache.hadoop.conf.Configuration +import org.apache.spark.{Partition, SerializableWritable, SparkContext, TaskContext} +import org.apache.spark.rdd.RDD +import org.apache.spark.sql.avro.{AvroDeserializer, AvroSerializer} +import org.apache.spark.sql.catalyst.InternalRow +import org.apache.spark.sql.catalyst.expressions.{SpecificInternalRow, UnsafeProjection, UnsafeRow} +import org.apache.spark.sql.execution.datasources.PartitionedFile +import org.apache.spark.sql.vectorized.ColumnarBatch + +import scala.collection.JavaConverters._ +import scala.collection.mutable +import scala.util.Try + +case class HoodieMergeOnReadPartition(index: Int, split: HoodieMergeOnReadFileSplit) extends Partition + +class HoodieMergeOnReadRDD(@transient sc: SparkContext, + @transient config: Configuration, + fullSchemaFileReader: PartitionedFile => Iterator[Any], + requiredSchemaFileReader: PartitionedFile => Iterator[Any], + tableState: HoodieMergeOnReadTableState) + extends RDD[InternalRow](sc, Nil) { + + private val confBroadcast = sc.broadcast(new SerializableWritable(config)) + + override def compute(split: Partition, context: TaskContext): Iterator[InternalRow] = { + val mergeParquetPartition = split.asInstanceOf[HoodieMergeOnReadPartition] + mergeParquetPartition.split match { + case dataFileOnlySplit if dataFileOnlySplit.logPaths.isEmpty => + read(mergeParquetPartition.split.dataFile, requiredSchemaFileReader) + case skipMergeSplit if skipMergeSplit.mergeType + .equals(DataSourceReadOptions.REALTIME_SKIP_MERGE_OPT_VAL) => + skipMergeFileIterator( + skipMergeSplit, + read(mergeParquetPartition.split.dataFile, requiredSchemaFileReader), + getConfig + ) + case payloadCombineSplit if payloadCombineSplit.mergeType + .equals(DataSourceReadOptions.REALTIME_PAYLOAD_COMBINE_OPT_VAL) => + payloadCombineFileIterator( + payloadCombineSplit, + read(mergeParquetPartition.split.dataFile, fullSchemaFileReader), Review comment: Yes there is still some room for improvement. Spark schema is tricky, when it passed the required columns into the `PrunedFilterScan`, the order of the columns are different from the actual schema, and it will get reorder in somewhere else before return to the user. The projected InternalRow and the projected Avro record will have a different order of schema. I will look into this but not sure if I am able to get it into this release. ---------------------------------------------------------------- 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: [email protected]
