第 5 步：实施使用者

通过实例化大小正确的 processRecord 对象并将记录数据添加到该对象来实施 StockTrade 方法，并在出现问题时记录警告。

byte[] arr = new byte[record.data().remaining()];
record.data().get(arr);
StockTrade trade = StockTrade.fromJsonAsBytes(arr);
    if (trade == null) {
        log.warn("Skipping record. Unable to parse record into StockTrade. Partition Key: " + record.partitionKey());
        return;
        }
stockStats.addStockTrade(trade);
                

实施简单的 reportStats 方法。可随时将输出格式修改为适合您的首选格式。

System.out.println("****** Shard " + kinesisShardId + " stats for last 1 minute ******\n" +
stockStats + "\n" +
"****************************************************************\n");
                

实施 resetStats 方法，这将创建新的 stockStats 实例。

stockStats = new StockStats();
                

实施 ShardRecordProcessor 接口所需的以下方法

@Override
public void leaseLost(LeaseLostInput leaseLostInput) {
    log.info("Lost lease, so terminating.");
}

@Override
public void shardEnded(ShardEndedInput shardEndedInput) {
    try {
        log.info("Reached shard end checkpointing.");
        shardEndedInput.checkpointer().checkpoint();
    } catch (ShutdownException | InvalidStateException e) {
        log.error("Exception while checkpointing at shard end. Giving up.", e);
    }
}

@Override
public void shutdownRequested(ShutdownRequestedInput shutdownRequestedInput) {
    log.info("Scheduler is shutting down, checkpointing.");
    checkpoint(shutdownRequestedInput.checkpointer());
}

private void checkpoint(RecordProcessorCheckpointer checkpointer) {
    log.info("Checkpointing shard " + kinesisShardId);
    try {
        checkpointer.checkpoint();
    } catch (ShutdownException se) {
        // Ignore checkpoint if the processor instance has been shutdown (fail over).
        log.info("Caught shutdown exception, skipping checkpoint.", se);
    } catch (ThrottlingException e) {
        // Skip checkpoint when throttled. In practice, consider a backoff and retry policy.
        log.error("Caught throttling exception, skipping checkpoint.", e);
    } catch (InvalidStateException e) {
        // This indicates an issue with the DynamoDB table (check for table, provisioned IOPS).
        log.error("Cannot save checkpoint to the DynamoDB table used by the Amazon Kinesis Client Library.", e);
    }
}
                

运行您在 第 4 步：实施创建器 中编写的创建者以将模拟股票交易记录引入流中。

验证之前（在创建 IAM 用户时）检索到的访问密钥和私有密钥对是否保存到文件 ~/.aws/credentials 中。

使用以下参数运行 StockTradesProcessor 类：

StockTradesProcessor StockTradeStream us-west-2

请注意，如果您在 us-west-2 之外的区域中创建流，则必须改为在此处指定该区域。