SQL access via Spark does not work

[Marco]

Hi,

I'm just trying out Apache Ignite and have issues when trying to access the
cache by using  SQL in Spark.

I have a simple data container defined:

case class DataContainer(@QuerySqlField key: Int, @QuerySqlField value:
String) extends Serializable{

  override def toString = s"Key$key, value $value"

}

Then I put some data into the cache via a Spark Batch app:
    val dataContainerArray = Array[DataContainer](new DataContainer(1,
"T"), new DataContainer(2, "A"))
    val dataContainerRdd = sc.parallelize(dataContainerArray)
    val dataKeyValueRdd = dataContainerRdd.map(item => Pair(item.key, item))

    //configure the cache...
    val ccfg = new CacheConfiguration[Int, DataContainer]()
    ccfg.setIndexedTypes(Int.getClass, DataContainer.getClass)
    ccfg.setName("C2")

    val cacheRdd = igniteContext.fromCache(ccfg)
    cacheRdd.savePairs(dataKeyValueRdd, true)


Now, I am trying to read the cache from another Spark app

val igniteContext = new IgniteContext[Int, DataContainer](sc,
      () => new IgniteConfiguration())

    //get the cache as a rdd representation
    val cacheRdd = igniteContext.fromCache("C2")

    // get the entry for key = 1 (WORKS FINE)
    val result = cacheRdd.filter(_._1 == 1).collect()
    result.foreach(row => println(row._2))


    //no result :-(
    cacheRdd.sql("select * from
C2.DATACONTAINER_").foreach(row=>println(row.toString()))

The access by the filter function(RDD means) works fine. Unfortunately, the
SQL access does not return any results. Does anybody have an idea what I am
making wrong here ?


Thanks,
Marco