使用SparkSQL刪除MySQL表

Question

我知道我們可以通過以下命令從SparkSQL查詢或創建Mysql表。

val data = sqlContext.read.jdbc(urlstring, tablename, properties)
data.write.format("com.databricks.spark.csv").save(result_location)

val dataframe = sqlContext.read.json("users.json")
dataframe.write.jdbc(urlstring, table, properties)

像那樣有什么方法可以放桌子嗎？

Answer 1

您可以嘗試使用JDBC驅動程序進行基本的DROP操作：

val DB_URL: String = ???
val USER: String = ???
val PASS: String = ???

def dropTable(tableName: String) = {

    import java.sql._;

    var conn: Connection = null;
    var stmt: Statement = null;

    try {
      Class.forName("com.mysql.jdbc.Driver");
      println("Connecting to a selected database...");
      conn = DriverManager.getConnection(DB_URL, USER, PASS);
      println("Connected database successfully...");
      println("Deleting table in given database...");
      stmt = conn.createStatement();
      val sql: String = s"DROP TABLE ${tableName} ";
      stmt.executeUpdate(sql);
      println(s"Table ${tableName} deleted in given database...");
    } catch {
      case e: Exception => println("exception caught: " + e);
    } finally {
      ???
    }
}

dropTable("test")

您可以使用JDBCUtils在Spark上JDBCUtils此操作，但這非常簡單。