[
https://issues.apache.org/jira/browse/PHOENIX-6783?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17877768#comment-17877768
]
ASF GitHub Bot commented on PHOENIX-6783:
-----------------------------------------
rejeb commented on code in PR #139:
URL:
https://github.com/apache/phoenix-connectors/pull/139#discussion_r1736647542
##########
phoenix5-spark/src/main/scala/org/apache/phoenix/spark/datasource/v2/PhoenixSparkSqlRelation.scala:
##########
@@ -0,0 +1,58 @@
+package org.apache.phoenix.spark.datasource.v2
+
+import org.apache.phoenix.spark.datasource.v2.reader.PhoenixDataSourceReader
+import org.apache.spark.rdd.RDD
+import org.apache.spark.sql.{DataFrame, Row, SQLContext, SaveMode,
SparkSession}
+import org.apache.spark.sql.catalyst.expressions.GenericRowWithSchema
+import org.apache.spark.sql.execution.datasources.v2.DataSourceRDD
+import org.apache.spark.sql.sources.v2.DataSourceOptions
+import org.apache.spark.sql.sources.{BaseRelation, Filter, InsertableRelation,
PrunedFilteredScan}
+import org.apache.spark.sql.types.StructType
+
+import scala.collection.JavaConverters._
+
+case class PhoenixSparkSqlRelation(
+ @transient sparkSession: SparkSession,
+ params: Map[String, String]
+ ) extends BaseRelation with
PrunedFilteredScan with InsertableRelation {
+
+ override def schema: StructType = dataSourceReader.readSchema()
+
+ override def sqlContext: SQLContext = sparkSession.sqlContext
+
+ private def dataSourceReader: PhoenixDataSourceReader = new
PhoenixDataSourceReader(dataSourceOptions)
+
+ private def dataSourceOptions = new DataSourceOptions(params.asJava)
+
+ override def buildScan(requiredColumns: Array[String], filters:
Array[Filter]): RDD[Row] = {
+ val requiredSchema = StructType(requiredColumns.flatMap(c =>
schema.fields.find(_.name == c)))
+
+ val reader: PhoenixDataSourceReader = dataSourceReader
+ reader.pushFilters(filters)
+ reader.pruneColumns(requiredSchema)
+ val rdd = new DataSourceRDD(
+ sqlContext.sparkContext,
+ reader.planInputPartitions().asScala
+ )
+ rdd.map(ir => {
+ new GenericRowWithSchema(ir.toSeq(requiredSchema).toArray,
requiredSchema)
+ })
+ }
+
+
+ override def insert(data: DataFrame, overwrite: Boolean): Unit = {
+ data
+ .write
+ .format("phoenix")
+ .option(PhoenixDataSource.TABLE, params(PhoenixDataSource.TABLE))
+ .option(PhoenixDataSource.JDBC_URL,
PhoenixDataSource.getJdbcUrlFromOptions(dataSourceOptions))
Review Comment:
The PhoenixDataSource.getJdbcUrlFromOptions method handle passing jdbcUrl,
zkUrl and also when none is specified.
> Phoenix5-Spark3 Connector Spark SQL Support
> -------------------------------------------
>
> Key: PHOENIX-6783
> URL: https://issues.apache.org/jira/browse/PHOENIX-6783
> Project: Phoenix
> Issue Type: Bug
> Components: connectors, spark-connector
> Environment: # Pure Spark 3.0.3
> # phoenix5-spark3-shaded-6.0.0-SNAPSHOT.jar located in $SPARK_HOME/jars
> # SQL used to create Spark Phoenix data source table is:
> {{create table Table1 (ID long, COL1 string) using phoenix options (table
> 'Table1', zkUrl '192.168.0.103:2181', primary 'ID')}}
> Reporter: yj
> Assignee: rejeb ben rejeb
> Priority: Major
>
> While conducting the Spark 3.0 Integration test using *Phoenix5-spark3-shaded
> module of project
> [Phoenix-Connectors|https://github.com/apache/phoenix-connectors]* the
> following questions occurred.
>
> The test process is as follows:
> 1. Phoenix created a table called "TABLE1" and then loaded the data into that
> table.
> 2. After that, I tried to select the data of the table using Spark.
>
> As written in the README of Phoenix5-spark3 module, the part that loads
> Phoenix data into the Spark DataFrame using the Phoenix data source seems to
> work well.
> However, when I *create Phoenix Spark data source table through Spark SQL*
> and try to select that data source table, the following error occurs:
> {{java.util.concurrent.ExecutionException:
> org.apache.spark.sql.AnalysisException: phoenix is not a valid Spark SQL Data
> Source}}
>
> I wonder if the Phoenix-Connectors does not support creating Phoenix Spark
> data source tables, or if there is any other reason for this error.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
