当前位置：首页 > article >正文

java Kafka批量消费和单个消费消息

article 2025/3/3 17:55:56

1、批量消费消息

1.1 配置参数

fetch.min.bytes：拉取的最小字节数

fetch.max.bytes：拉取的最大字节数

max.partition.fetch.bytes：分区拉取的最大字节数

fetch.max.wait.ms：拉取操作的最大等待时间

max.poll.records：拉取操作最大获取的记录数量

以上参数决定了一次消费能消费多少条消息。

    @Bean
    public KafkaListenerContainerFactory<?> kafkaBatchConsumerFactory() {
        return batchFactory(consumerConfig(KafkaConsumerGroupIdConstant.KAFKA_CONSUMER_BATCH_GROUP, KafkaConsumerResetOffsetEnum.EARLIEST.getType()));
    }

1.2 消费代码

    @KafkaListener(groupId = KafkaConsumerGroupIdConstant.KAFKA_CONSUMER_BATCH_GROUP,
            topics = KafkaTopicConstant.KAFKA_TOPIC,
            containerFactory = "kafkaBatchConsumerFactory")
    public void batchConsume(List<ConsumerRecord<String, String>> records, Acknowledgment ack) {
        String topicName = KafkaTopicConstant.KAFKA_TOPIC;
        try {
            List<String> valueList = records.stream()
                    .map(ConsumerRecord::value).collect(Collectors.toList());
            for (String value : valueList) {
                log.info("批量消费topic:{}, value:{}", topicName, value);
            }
            ack.acknowledge();
        } catch (Exception e) {
            log.error("kafka消费{}:", topicName, e);
        }
    }

1.3 消费结果

2、单个消费消息

2.1 配置参数

    @Bean
    public KafkaListenerContainerFactory<?> kafkaSingleConsumerFactory() {
        ConcurrentKafkaListenerContainerFactory<Integer, String> resultFactory = batchFactory(consumerConfig(KafkaConsumerGroupIdConstant.KAFKA_CONSUMER_SINGLE_GROUP, KafkaConsumerResetOffsetEnum.EARLIEST.getType()));
        resultFactory.setBatchListener(false);
        return resultFactory;
    }

2.2 消费代码


    @KafkaListener(groupId = KafkaConsumerGroupIdConstant.KAFKA_CONSUMER_SINGLE_GROUP,
            topics = KafkaTopicConstant.KAFKA_TOPIC,
            containerFactory = "kafkaSingleConsumerFactory")
    public void singleConsume(ConsumerRecord<String, String> records, Acknowledgment ack) {
        String topicName = KafkaTopicConstant.KAFKA_TOPIC;
        try {
            log.info("单个消费 topic:{}, value:{}", topicName, records.value());
            ack.acknowledge();
        } catch (Exception e) {
            log.error("kafka消费{}:", topicName, e);
        }
    }

2.3 消费结果

注：

部分代码未能在博客内体现，请点击以下链接跳转至Gitee的xudongbase项目的kafka分支。

xudongbase: 主要是项目中可以用到的共通方法，现有easyexcel分支在持续更新中。欢迎大家Star和提交Issues。easyexcel分支：批量设置样式，批量添加批注，批量合并单元格，设置冻结行和列，设置行高列宽，隐藏行和列，绑定下拉框数据，设置水印，插入图片 - Gitee.comhttps://gitee.com/xudong_master/xudongbase/tree/kafka/