Spark SQL 高并发读写数据库

//Construct a DataFrame representing the database table accessible via JDBC URL url named table and connection properties.
Dataset<Row> 	jdbc(String url, String table, java.util.Properties properties)
//Construct a DataFrame representing the database table accessible via JDBC URL url named table using connection properties.
Dataset<Row> 	jdbc(String url, String table, String[] predicates, java.util.Properties connectionProperties)
//Construct a DataFrame representing the database table accessible via JDBC URL url named table.
Dataset<Row> 	jdbc(String url, String table, String columnName, long lowerBound, long upperBound, int numPartitions, java.util.Properties connectionProperties)

SparkSession sparkSession = SparkSession.builder().appName("SPARK_FENGDING_TASK1").master("local").config("spark.testing.memory", 471859200).getOrCreate();
// 配置连接属性
Properties dbProps = new Properties();
dbProps.put("user","user");
dbProps.put("password","pwd");
dbProps.put("driver","oracle.jdbc.driver.OracleDriver");
// 连接数据库 获取数据 要使用自己的数据库连接串
Dataset<Row> tableDf = sparkSession.read().jdbc("jdbc:oracle:thin:@IP:1521:DEMO", "TABLE_DEMO", dbProps);
// 返回1
tableDf.rdd().getPartitions();

def jdbc(
    url: String,
    table: String,
    predicates: Array[String], #这个是分层的条件，一个数组
    connectionProperties: Properties): DataFrame = {
    val parts: Array[Partition] = predicates.zipWithIndex.map { case (part, i) =>
        JDBCPartition(part, i) : Partition
    }
    jdbc(url, table, parts, connectionProperties)
}

// 设置分区条件 通过入库时间 把 10月和11月 的数据 分两个分区
String[] patitions = {"rksj >= '1569859200' and rksj < '1572537600'","rksj >= '1572537600' and rksj < '1575129600'"};
// 根据StudentId 分15个分区，就会有15个task抽取数据
Dataset<Row> tableDf3 = sparkSession.read().jdbc("jdbc:oracle:thin:@IP:1521:DEMO", "TABLE_DEMO",patitions,dbProps);
// 返回2
tableDf3.rdd().getPartitions();

def jdbc(
    url: String,
    table: String,
    columnName: String,    # 根据该字段分区，需要为整形，比如id等
    lowerBound: Long,      # 分区的下界
    upperBound: Long,      # 分区的上界
    numPartitions: Int,    # 分区的个数
    connectionProperties: Properties): DataFrame = {
    val partitioning = JDBCPartitioningInfo(columnName, lowerBound, upperBound, numPartitions)
    val parts = JDBCRelation.columnPartition(partitioning)
    jdbc(url, table, parts, connectionProperties)
}

// 根据StudentId 分15个分区，就会有15个task抽取数据
Dataset<Row> tableDf2 = sparkSession.read().jdbc("jdbc:oracle:thin:@IP:1521:DEMO", "TABLE_DEMO", "studentId",0,1500,15,dbProps);
// 返回10
tableDf2.rdd().getPartitions();

// 第一张写法，指定format类型，使用save方法存储数据库
jdbcDF.write()
  .format("jdbc")
  .option("url", "jdbc:postgresql:dbserver")
  .option("dbtable", "schema.tablename")
  .option("user", "username")
  .option("password", "password")
  .save();
// 第二种写法 使用jdbc写入数据库
jdbcDF2.write()
  .jdbc("jdbc:postgresql:dbserver", "schema.tablename", connectionProperties);

// 第三种写法,也是使用jdbc，只不过添加createTableColumnTypes，创建表的时候使用该属性字段创建表字段
jdbcDF.write()
  .option("createTableColumnTypes", "name CHAR(64), comments VARCHAR(1024)")
  .jdbc("jdbc:postgresql:dbserver", "schema.tablename", connectionProperties);