Re: Operation not allowed: STORED BY (from Spark Dataframe save)

[yannv]

Hello, 

I am trying to create a carbon data table from a Spark Data Frame, however I
am getting an error with the (automatic create table statement) 

I run this code on spark-shell (passing the carbon data assembly jar file
for 1.4.0 as well as master branch), on Azure HDInsight cluster with spark
2.2.1. 

Code used : 


import java.io.File

import org.apache.spark.sql.{SaveMode, SparkSession}

import org.apache.carbondata.core.constants.CarbonCommonConstants
import org.apache.carbondata.core.util.CarbonProperties

import org.apache.spark.sql.CarbonSession._

import spark.implicits._

val rootPath = new
File(this.getClass.getResource("/").getPath).getCanonicalPath
val storeLocation = s"$rootPath/store"
val warehouse = s"$rootPath/warehouse"
val metastoredb = s"$rootPath/metastore"

CarbonProperties.getInstance()
  .addProperty(CarbonCommonConstants.CARBON_TIMESTAMP_FORMAT, "yyyy/MM/dd
HH:mm:ss")
  .addProperty(CarbonCommonConstants.CARBON_DATE_FORMAT, "yyyy/MM/dd")
  .addProperty(CarbonCommonConstants.ENABLE_UNSAFE_COLUMN_PAGE, "true")
  .addProperty(CarbonCommonConstants.CARBON_BADRECORDS_LOC, "")


val carbon = SparkSession
  .builder()
  .config("spark.sql.warehouse.dir", warehouse)
  .config("spark.sql.crossJoin.enabled", "true")
  .getOrCreateCarbonSession(storeLocation, metastoredb)


val df = carbon.sparkContext.parallelize(1 to 50)
  .map(x => ("c1" + x % 10, "c2", x))
  .toDF("col1", "col2", "num")

df.write
      .format("carbondata")
      .option("tableName", "carbon_table")
      .option("partitionColumns", "col1")
      .mode(SaveMode.Overwrite)
      .save()



This is the error I am getting : 

org.apache.spark.sql.catalyst.parser.ParseException:
Operation not allowed: STORED BY(line 5, pos 1)

== SQL ==

 CREATE TABLE IF NOT EXISTS default.carbon_df_table_test1
 (c2 STRING, number INT)
 PARTITIONED BY (c1 string)
 STORED BY 'carbondata'
-^^^

  TBLPROPERTIES ('STREAMING' = 'false')



  at
org.apache.spark.sql.catalyst.parser.ParserUtils$.operationNotAllowed(ParserUtils.scala:39)
  at
org.apache.spark.sql.execution.SparkSqlAstBuilder$$anonfun$visitCreateFileFormat$1.apply(SparkSqlParser.scala:1194)
  at
org.apache.spark.sql.execution.SparkSqlAstBuilder$$anonfun$visitCreateFileFormat$1.apply(SparkSqlParser.scala:1186)
  at
org.apache.spark.sql.catalyst.parser.ParserUtils$.withOrigin(ParserUtils.scala:99)
  at
org.apache.spark.sql.execution.SparkSqlAstBuilder.visitCreateFileFormat(SparkSqlParser.scala:1185)
  at
org.apache.spark.sql.execution.SparkSqlAstBuilder$$anonfun$visitCreateHiveTable$1$$anonfun$31.apply(SparkSqlParser.scala:1090)
  at
org.apache.spark.sql.execution.SparkSqlAstBuilder$$anonfun$visitCreateHiveTable$1$$anonfun$31.apply(SparkSqlParser.scala:1090)
  at scala.Option.map(Option.scala:146)
  at
org.apache.spark.sql.execution.SparkSqlAstBuilder$$anonfun$visitCreateHiveTable$1.apply(SparkSqlParser.scala:1090)



I tried various constructors for the carbon object without success. 
Note : I can create a Carbondata table and insert data from CSV file
successfully, but it looks like when the save is executed it tries to create
the (new) table and I get this error on "Stored by"... 

carbon.sql("create table IF NOT EXISTS carbon_test_csv(id int, name string,
scale decimal, country string, salary double) STORED BY 'carbondata'")
carbon.sql("LOAD DATA INPATH '/tmp/sample.csv' INTO TABLE
hive_carbon_test_csv")

Regards, 
Yann





--
Sent from: 
http://apache-carbondata-dev-mailing-list-archive.1130556.n5.nabble.com/