zhenlineo commented on code in PR #40025: URL: https://github.com/apache/spark/pull/40025#discussion_r1107634661
########## connector/connect/client/jvm/src/main/scala/org/apache/spark/sql/DataFrameReader.scala: ########## @@ -0,0 +1,408 @@ +/* + * 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 scala.collection.JavaConverters._ + +import org.apache.spark.annotation.Stable +import org.apache.spark.internal.Logging +import org.apache.spark.sql.catalyst.util.{CaseInsensitiveMap, CharVarcharUtils} +import org.apache.spark.sql.types.StructType + +/** + * Interface used to load a [[Dataset]] from external storage systems (e.g. file systems, + * key-value stores, etc). Use `SparkSession.read` to access this. + * + * @since 3.4.0 + */ +@Stable +class DataFrameReader private[sql](sparkSession: SparkSession) extends Logging { + + /** + * Specifies the input data source format. + * + * @since 3.4.0 + */ + def format(source: String): DataFrameReader = { + this.source = source + this + } + + /** + * Specifies the input schema. Some data sources (e.g. JSON) can infer the input schema + * automatically from data. By specifying the schema here, the underlying data source can + * skip the schema inference step, and thus speed up data loading. + * + * @since 3.4.0 + */ + def schema(schema: StructType): DataFrameReader = { + if (schema != null) { + val replaced = CharVarcharUtils.failIfHasCharVarchar(schema).asInstanceOf[StructType] + this.userSpecifiedSchema = Option(replaced) + } + this + } + + /** + * Specifies the schema by using the input DDL-formatted string. Some data sources (e.g. JSON) can + * infer the input schema automatically from data. By specifying the schema here, the underlying + * data source can skip the schema inference step, and thus speed up data loading. + * + * {{{ + * spark.read.schema("a INT, b STRING, c DOUBLE").csv("test.csv") + * }}} + * + * @since 3.4.0 + */ + def schema(schemaString: String): DataFrameReader = { + schema(StructType.fromDDL(schemaString)) + } + + /** + * Adds an input option for the underlying data source. + * + * All options are maintained in a case-insensitive way in terms of key names. + * If a new option has the same key case-insensitively, it will override the existing option. + * + * @since 3.4.0 + */ + def option(key: String, value: String): DataFrameReader = { + this.extraOptions = this.extraOptions + (key -> value) + this + } + + /** + * Adds an input option for the underlying data source. + * + * All options are maintained in a case-insensitive way in terms of key names. + * If a new option has the same key case-insensitively, it will override the existing option. + * + * @since 3.4.0 + */ + def option(key: String, value: Boolean): DataFrameReader = option(key, value.toString) + + /** + * Adds an input option for the underlying data source. + * + * All options are maintained in a case-insensitive way in terms of key names. + * If a new option has the same key case-insensitively, it will override the existing option. + * + * @since 3.4.0 + */ + def option(key: String, value: Long): DataFrameReader = option(key, value.toString) + + /** + * Adds an input option for the underlying data source. + * + * All options are maintained in a case-insensitive way in terms of key names. + * If a new option has the same key case-insensitively, it will override the existing option. + * + * @since 3.4.0 + */ + def option(key: String, value: Double): DataFrameReader = option(key, value.toString) + + /** + * (Scala-specific) Adds input options for the underlying data source. + * + * All options are maintained in a case-insensitive way in terms of key names. + * If a new option has the same key case-insensitively, it will override the existing option. + * + * @since 3.4.0 + */ + def options(options: scala.collection.Map[String, String]): DataFrameReader = { + this.extraOptions ++= options + this + } + + /** + * Adds input options for the underlying data source. + * + * All options are maintained in a case-insensitive way in terms of key names. + * If a new option has the same key case-insensitively, it will override the existing option. + * + * @since 3.4.0 + */ + def options(options: java.util.Map[String, String]): DataFrameReader = { + this.options(options.asScala) + this + } + + /** + * Loads input in as a `DataFrame`, for data sources that don't require a path (e.g. external + * key-value stores). + * + * @since 3.4.0 + */ + def load(): DataFrame = { + load(Seq.empty: _*) // force invocation of `load(...varargs...)` + } + + /** + * Loads input in as a `DataFrame`, for data sources that require a path (e.g. data backed by + * a local or distributed file system). + * + * @since 3.4.0 + */ + def load(path: String): DataFrame = { + // force invocation of `load(...varargs...)` + load(Seq(path): _*) + } + + /** + * Loads input in as a `DataFrame`, for data sources that support multiple paths. + * Only works if the source is a HadoopFsRelationProvider. + * + * @since 3.4.0 + */ + @scala.annotation.varargs + def load(paths: String*): DataFrame = { + sparkSession.newDataset { builder => + val dataSourceBuilder = builder.getReadBuilder.getDataSourceBuilder + assertSourceFormatSpecified() + dataSourceBuilder.setFormat(source) + userSpecifiedSchema.foreach(schema => dataSourceBuilder.setSchema(schema.toDDL)) + extraOptions.foreach { + case (k, v) => dataSourceBuilder.putOptions(k, v) + } + paths.foreach(path => dataSourceBuilder.addPaths(path)) Review Comment: @hvanhovell I do not think we can anything special here. The server planner will call the original SQL code and merge it based on the config settings. e.g. ``` def load(paths: String*): DataFrame = { .... val legacyPathOptionBehavior = sparkSession.sessionState.conf.legacyPathOptionBehavior if (!legacyPathOptionBehavior && (extraOptions.contains("path") || extraOptions.contains("paths")) && paths.nonEmpty) { throw QueryCompilationErrors.pathOptionNotSetCorrectlyWhenReadingError() } DataSource.lookupDataSourceV2(source, sparkSession.sessionState.conf).flatMap { provider => DataSourceV2Utils.loadV2Source(sparkSession, provider, userSpecifiedSchema, extraOptions, source, paths: _*) }.getOrElse(loadV1Source(paths: _*)) ``` If we merge them on the client we by default assumed the config value. So the best is actually leave the result to server SQL API to handle the merging of paths. -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
