xushiyan commented on a change in pull request #4611: URL: https://github.com/apache/hudi/pull/4611#discussion_r795138030
########## File path: hudi-spark-datasource/hudi-spark3/src/main/scala/org/apache/spark/sql/hudi/catalog/BaseStagedTable.scala ########## @@ -0,0 +1,57 @@ +/* + * 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.spark.sql.hudi.catalog + +import org.apache.hudi.exception.HoodieException +import org.apache.spark.sql.connector.catalog._ +import org.apache.spark.sql.connector.expressions.Transform +import org.apache.spark.sql.connector.write.{LogicalWriteInfo, WriteBuilder} +import org.apache.spark.sql.types.StructType + +import java.util + +/** + * Basic implementation that represents a table which is staged for being committed. + * @param ident table ident + * @param table table + * @param catalog table catalog + */ +case class BaseStagedTable(ident: Identifier, + table: Table, + catalog: TableCatalog) extends SupportsWrite with StagedTable { + override def newWriteBuilder(info: LogicalWriteInfo): WriteBuilder = { + info match { + case supportsWrite: SupportsWrite => supportsWrite.newWriteBuilder(info) + case _ => throw new HoodieException(s"Table implementation does not support writes: $ident.name") Review comment: should it still be `HoodieException` given this is not HoodieStagedTable? also suggested a clearer message. ```suggestion case _ => throw new HoodieException(s"Table `$ident.name` does not support writes.") ``` ########## File path: hudi-spark-datasource/hudi-spark3-common/src/main/scala/org/apache/spark/sql/HoodieSpark3SqlUtils.scala ########## @@ -0,0 +1,45 @@ +/* + * 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.spark.sql + +import org.apache.hudi.exception.HoodieException +import org.apache.spark.sql.catalyst.catalog.BucketSpec +import org.apache.spark.sql.connector.expressions.{BucketTransform, FieldReference, IdentityTransform, Transform} + +import scala.collection.mutable + +object HoodieSpark3SqlUtils { + def convertTransforms(partitions: Seq[Transform]): (Seq[String], Option[BucketSpec]) = { + val identityCols = new mutable.ArrayBuffer[String] + var bucketSpec = Option.empty[BucketSpec] + + partitions.map { + case IdentityTransform(FieldReference(Seq(col))) => + identityCols += col + + + case BucketTransform(numBuckets, FieldReference(Seq(col))) => + bucketSpec = Some(BucketSpec(numBuckets, col :: Nil, Nil)) + + case _ => + throw new HoodieException(s"Partitioning by expressions") Review comment: is this exception message user-facing ? then we need to clarify it further. ########## File path: hudi-spark-datasource/hudi-spark-common/src/main/scala/org/apache/hudi/IncrementalRelation.scala ########## @@ -171,7 +171,7 @@ class IncrementalRelation(val sqlContext: SQLContext, if (metaBootstrapFileIdToFullPath.nonEmpty) { df = sqlContext.sparkSession.read - .format("hudi") + .format("hudi_v1") Review comment: so incremental read still go through v1. is there any follow-up plan to use v2 for this as well ? ########## File path: hudi-spark-datasource/hudi-spark3/src/main/scala/org/apache/hudi/Spark3DefaultSource.scala ########## @@ -17,8 +17,30 @@ package org.apache.hudi +import org.apache.hudi.exception.HoodieException +import org.apache.spark.sql.SparkSession +import org.apache.spark.sql.connector.catalog.{Table, TableProvider} +import org.apache.spark.sql.connector.expressions.Transform +import org.apache.spark.sql.hudi.catalog.HoodieInternalV2Table import org.apache.spark.sql.sources.DataSourceRegister +import org.apache.spark.sql.types.StructType +import org.apache.spark.sql.util.CaseInsensitiveStringMap + +class Spark3DefaultSource extends DefaultSource with DataSourceRegister with TableProvider { -class Spark3DefaultSource extends DefaultSource with DataSourceRegister { override def shortName(): String = "hudi" + + def inferSchema: StructType = new StructType() + + override def inferSchema(options: CaseInsensitiveStringMap): StructType = inferSchema + + override def getTable(schema: StructType, + partitioning: Array[Transform], + properties: java.util.Map[String, String]): Table = { + val options = new CaseInsensitiveStringMap(properties) + val path = options.get("path") + if (path == null) throw new HoodieException("path is null") Review comment: is this also user-facing message? how about "'path' cannot be null; missing 'path' from table properties?" just an example. ########## File path: hudi-spark-datasource/hudi-spark3/src/main/scala/org/apache/spark/sql/hudi/catalog/BaseStagedTable.scala ########## @@ -0,0 +1,57 @@ +/* + * 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.spark.sql.hudi.catalog + +import org.apache.hudi.exception.HoodieException +import org.apache.spark.sql.connector.catalog._ +import org.apache.spark.sql.connector.expressions.Transform +import org.apache.spark.sql.connector.write.{LogicalWriteInfo, WriteBuilder} +import org.apache.spark.sql.types.StructType + +import java.util + +/** + * Basic implementation that represents a table which is staged for being committed. + * @param ident table ident + * @param table table + * @param catalog table catalog + */ +case class BaseStagedTable(ident: Identifier, Review comment: I was confused about the name and now think would it be better named `BasicStagedTable` or `SimpleStagedTable` ? `BaseXXX` implies to-be-extended but it's actually parallel to `HoodieStagedTable` ########## File path: hudi-spark-datasource/hudi-spark/src/main/scala/org/apache/spark/sql/hudi/HoodieSqlUtils.scala ########## @@ -19,10 +19,8 @@ package org.apache.spark.sql.hudi import org.apache.hudi.SparkAdapterSupport import org.apache.spark.sql.catalyst.TableIdentifier -import org.apache.spark.sql.catalyst.expressions.{And, Cast, Expression, Literal} +import org.apache.spark.sql.catalyst.expressions.{And, Expression} import org.apache.spark.sql.catalyst.plans.logical.{MergeIntoTable, SubqueryAlias} -import org.apache.spark.sql.internal.SQLConf -import org.apache.spark.sql.types.{DataType, NullType} object HoodieSqlUtils extends SparkAdapterSupport { Review comment: just wonder if this can be merged into `HoodieSqlCommonUtils` to have fewer classes for clarity -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
