我正在尝试使用本地spark集群的cdatajdbc驱动程序将数据插入azuredatabricks。我第一次尝试直接从这样的Dataframe

df.write
      .format("jdbc")
      .option("driver", DatabricksUtil.driver)
      .option("url", url)
      .option("dbtable", s"$schemaName.$tableName")
      .option("createTableColumnTypes", JdbcUtils.schemaString(df, url, None))
      .mode(SaveMode.Overwrite)
      .save()

甚至这个简单的查询

connection.createStatement.execute("CREATE TABLE `default`.`state2` (State Text, Abbreviation Text, Code Int)")

在这两种情况下，我都遇到了这个错误

Exception in thread "main" XcoreXsparksqlX190X7354.ezf: Error running query: org.apache.spark.sql.catalyst.parser.ParseException: 
DataType text is not supported.(line 1, pos 40)

== SQL ==
CREATE TABLE `default`.`state2`(`State` Text, `Abbreviation` Text, `Code` Int)
----------------------------------------^^^

    at XcoreXsparksqlX190X7354.vjf.a(Unknown Source)
    at XcoreXsparksqlX190X7354.nud.execute(Unknown Source)
    at c3.spark.util.DatabricksApp$.main(DatabricksUtil.scala:208)
    at c3.spark.util.DatabricksApp.main(DatabricksUtil.scala)

现在，如果我尝试使用sql语句或准备好的语句手动插入一些行，如下所示：

connection.createStatement.execute("DROP TABLE IF EXISTS `default`.`state2`")
    connection.createStatement.execute("CREATE TABLE `default`.`state2` (State String, Abbreviation String, Code Int)")
    // insert prepared statement
    val preparedStatement = connection.prepareStatement("INSERT INTO `default`.`state2` (State,Abbreviation,Code) VALUES (?,?,?)")
    preparedStatement.setString(1, "Alabama")
    preparedStatement.setString(2, "AL")
    preparedStatement.setInt(3, 1)
    preparedStatement.addBatch()
    preparedStatement.setString(1, "Alaska")
    preparedStatement.setString(2, "AK")
    preparedStatement.setInt(3, 2)
    preparedStatement.addBatch()
    println("Inserted rows: " + preparedStatement.executeBatch().mkString(","))
    // insert statement
    connection.createStatement().execute("INSERT INTO `default`.`state2` (State,Abbreviation,Code) VALUES (Arizona,AZ,3), (Arkansas,AR,4)")
    // read table
    val result = executeQuery("SELECT * FROM `default`.`state2`", connection.createStatement())
    val rows = Iterator.continually(result).takeWhile(_.next()).map{rs =>
      val row = Seq("String", "String", "Int").zipWithIndex.map{case (coltype, colindex) =>
        coltype match {
          case "String" => rs.getString(colindex + 1)
          case "Int" => rs.getInt(colindex + 1)
        }
      }
      row.mkString(",")
    }.toList
    println(rows)

我看到我们插入了一些行，但没有任何数据（字符串为null，整数为0）

Inserted rows: 1,1
List(null,null,0, null,null,0, null,null,0, null,null,0)

什么是正确的方式插入到数据库中？？？