教你如何让sparksql写mysql的时候支持update操作(2)-木庄网络博客

改造后的样子：

try {
      if (supportsTransactions) {
        conn.setAutoCommit(false) // Everything in the same db transaction.
        conn.setTransactionIsolation(finalIsolationLevel)
      }
//      val stmt = insertStatement(conn, table, rddSchema, dialect)
      //此处采用最新自己的sql语句，封装成prepareStatement
      val stmt = conn.prepareStatement(sqlStmt)
      println(sqlStmt)
      /**
        * 在mysql中有这样的操作：
        * INSERT INTO user_admin_t (_id,password) VALUES ('1','第一次插入的密码')
        * INSERT INTO user_admin_t (_id,password）VALUES ('1','第一次插入的密码') ON DUPLICATE KEY UPDATE _id = 'UpId',password = 'upPassword';
        * 如果是下面的ON DUPLICATE KEY操作，那么在prepareStatement中的游标会扩增一倍
        * 并且如果没有update操作，那么他的游标是从0开始计数的
        * 如果是update操作，要算上之前的insert操作
        * */
        //makeSetter也要适配update操作，即游标问题
?
      val isUpdate = saveMode == CustomSaveMode.Update
      val setters: Array[JDBCValueSetter] = isUpdate match {
        case true =>
          val setters: Array[JDBCValueSetter] = rddSchema.fields.map(_.dataType)
            .map(makeSetter(conn, dialect, _)).toArray
          Array.fill(2)(setters).flatten
        case _ =>
          rddSchema.fields.map(_.dataType)
      val numFieldsLength = rddSchema.fields.length
      val numFields = isUpdate match{
        case true => numFieldsLength *2
        case _ => numFieldsLength
      val cursorBegin = numFields / 2
      try {
        var rowCount = 0
        while (iterator.hasNext) {
          val row = iterator.next()
          var i = 0
          while (i < numFields) {
            if(isUpdate){
              //需要判断当前游标是否走到了ON DUPLICATE KEY UPDATE
              i < cursorBegin match{
                  //说明还没走到update阶段
                case true =>
                  //row.isNullAt 判空,则设置空值
                  if (row.isNullAt(i)) {
                    stmt.setNull(i + 1, nullTypes(i))
                  } else {
                    setters(i).apply(stmt, row, i, 0)
                  }
                  //说明走到了update阶段
                case false =>
                  if (row.isNullAt(i - cursorBegin)) {
                    //pos - offset
                    stmt.setNull(i + 1, nullTypes(i - cursorBegin))
                    setters(i).apply(stmt, row, i, cursorBegin)
              }
            }else{
              if (row.isNullAt(i)) {
                stmt.setNull(i + 1, nullTypes(i))
              } else {
                setters(i).apply(stmt, row, i ,0)
            }
            //滚动游标
            i = i + 1
          }
          stmt.addBatch()
          rowCount += 1
          if (rowCount % batchSize == 0) {
            stmt.executeBatch()
            rowCount = 0
        }
        if (rowCount > 0) {
          stmt.executeBatch()
      } finally {
        stmt.close()
        conn.commit()
      committed = true
      Iterator.empty
    } catch {
      case e: SQLException =>
        val cause = e.getNextException
        if (cause != null && e.getCause != cause) {
          if (e.getCause == null) {
            e.initCause(cause)
          } else {
            e.addSuppressed(cause)
        throw e
    } finally {
      if (!committed) {
        // The stage must fail.  We got here through an exception path, so
        // let the exception through unless rollback() or close() want to
        // tell the user about another problem.
        if (supportsTransactions) {
          conn.rollback()
        conn.close()
      } else {
        // The stage must succeed.  We cannot propagate any exception close() might throw.
        try {
          conn.close()
        } catch {
          case e: Exception => logWarning("Transaction succeeded, but closing failed", e)

// A `JDBCValueSetter` is responsible for setting a value from `Row` into a field for
  // `PreparedStatement`. The last argument `Int` means the index for the value to be set
  // in the SQL statement and also used for the value in `Row`.
  //PreparedStatement, Row, position , cursor
  private type JDBCValueSetter = (PreparedStatement, Row, Int , Int) => Unit
?
  private def makeSetter(
      conn: Connection,
      dialect: JdbcDialect,
      dataType: DataType): JDBCValueSetter = dataType match {
    case IntegerType =>
      (stmt: PreparedStatement, row: Row, pos: Int,cursor:Int) =>
        stmt.setInt(pos + 1, row.getInt(pos - cursor))
    case LongType =>
        stmt.setLong(pos + 1, row.getLong(pos - cursor))
    case DoubleType =>
        stmt.setDouble(pos + 1, row.getDouble(pos - cursor))
    case FloatType =>
        stmt.setFloat(pos + 1, row.getFloat(pos - cursor))
    case ShortType =>
        stmt.setInt(pos + 1, row.getShort(pos - cursor))
    case ByteType =>
        stmt.setInt(pos + 1, row.getByte(pos - cursor))
    case BooleanType =>
        stmt.setBoolean(pos + 1, row.getBoolean(pos - cursor))
    case StringType =>
//        println(row.getString(pos))
        stmt.setString(pos + 1, row.getString(pos - cursor))
    case BinaryType =>
        stmt.setBytes(pos + 1, row.getAs[Array[Byte]](pos - cursor))
    case TimestampType =>
        stmt.setTimestamp(pos + 1, row.getAs[java.sql.Timestamp](pos - cursor))
    case DateType =>
        stmt.setDate(pos + 1, row.getAs[java.sql.Date](pos - cursor))
    case t: DecimalType =>
        stmt.setBigDecimal(pos + 1, row.getDecimal(pos - cursor))
    case ArrayType(et, _) =>
      // remove type length parameters from end of type name
      val typeName = getJdbcType(et, dialect).databaseTypeDefinition
        .toLowerCase.split("\\(")(0)
        val array = conn.createArrayOf(
          typeName,
          row.getSeq[AnyRef](pos - cursor).toArray)
        stmt.setArray(pos + 1, array)
    case _ =>
      (_: PreparedStatement, _: Row, pos: Int,cursor:Int) =>
        throw new IllegalArgumentException(
          s"Can't translate non-null value for field $pos")
  }

完整代码：

https://github.com/niutaofan/bazinga

到此这篇关于教你如何让spark sql写mysql的时候支持update操作的文章就介绍到这了,更多相关spark sql写mysql支持update内容请搜索

标签：SQL

返回前面的内容