博客

1.如何在项目中使用kafka?

1.1)因为kafka的使用依赖于zookeeper(https://mp.weixin.qq.com/s/geR3pDw_Yjhmu8KMsXQosg在kafka v2.8版本后将zookeeper也集成在了服务中在kafka v2.8版本后官网取消了kafka依赖zookeeper集群的机制,采用内置kraft的方式),配置zookeeper的信息,即需要在kafka/config/zookeeper.properties配置zookeeper服务运行期间产生的数据存放位置dataDir,

 1.2)在kafka/config/server.properties 中配置kafka服务运行期间产生的log文件的位置，(注意:Kafka 使用消息日志（Log）来保存数据)

1.3)运行zookeeper和kafka服务,

./bin/zookeeper-server-start.sh ./config/zookeeper.properties
./bin/kafka-server-start.sh ./config/server.properties

 1.4)在spring中引入kafka依赖

<dependency>
     <groupId>org.springframework.kafka</groupId>
     <artifactId>spring-kafka</artifactId>
</dependency>

spring中如何使用kafka?

发送消息：利用KafkaTemplate类->kafkaTemplate.send(topic, message);

消费消息：利用注解@KafkaListener(topics = {**},groupId = “”)

1.5)应用：如何在项目中使用kafka实现发布系统通知功能？

用户进行关注,评论,回复等行为都是时刻在发生的.如果当某个用户进行关注/评论/回复(上游服务)后,系统立即对此请求进行处理，即向被关注/评论/回复的用户发送系统通知(下游服务)这一功能,当用户关注,评论,回复行为异常活跃时,就会拖垮服务器或者数据库.

利用kafka消息引擎的异步,解耦,流量削峰的特性,来实现项目中系统通知的功能.

1.5.1)将用户进行关注/评论/回复的行为抽象为事件Event对象

@Component
public class Event {
    String topic;//事件的主题(点赞/关注/回复)
    int userId;//事件的发起者
    int entityType;//被点赞/关注/回复的实体类型(项目中只有三种:用户,帖子,评论)
    int entityId;//被点赞/关注/回复的实体id
    int entityUserId;//被点赞/关注/回复的实体的作者（实体类型为用户时，entityId==entityUserId）
    Map<String, Object> data = new HashMap<>();//事件中其他额外需要装的数据

    /**
    省略get/set方法
    */
}

1.5.2)写生产者Producer类

@Component
public class EventProducer {
    @Autowired
    private KafkaTemplate kafkaTemplate;

    public void fireEvent(Event event){
        kafkaTemplate.send(event.getTopic(), JSONObject.toJSONString(event));
    }
}

1.5.3)写消费者Consumer 类

@Component
public class EventConsumer implements CommunityConst {


    @Autowired
    private MessageService messageService;


    @KafkaListener(topics = {KAFKA_TOPIC_COMMENT,KAFKA_TOPIC_LIKE,KAFKA_TOPIC_FOLLOW})
    public void handleCommentMessage(ConsumerRecord record){
//spring监听到以上某一个主题下有消息了,就会自动调用此方法,并将消息封装到ConsumerRecord对象中
        if(record ==null || record.value()!=null) {
            logger.error("event is null");
        }
        Event event = JSONObject.parseObject( record.value().toString(), Event.class);
        if(event == null){
            logger.error("event is not null, but it has a wrong form");
        }
        //produce message
        Message message = new Message();
        message.setFromId(SYSTEM_USER_ID);
        message.setToId(event.getEntityUserId());
        message.setConversationId(event.getTopic());
        message.setCreateTime(new Date());

        Map<String, Object> content = new HashMap<>();
        content.put("userId",event.getUserId());
        content.put("entityType",event.getEntityType());
        content.put("entityId",event.getEntityId());
        if(event.getData()!=null){
            for(Map.Entry<String, Object> entry: event.getData().entrySet()){
                content.put(entry.getKey(),entry.getValue());
            }
        }
        message.setContent(JSONObject.toJSONString(content));
        //add a system message
       messageService.sendMessage(message);//插入一条消息到数据库消息表中
      
    }

}

1.5.4)什么时刻触发消息的发送

以评论行为为例，

@Controller
@RequestMapping(path = "/comment")
public class CommentController implements CommunityConst {
    @Autowired
    private HostHolder hostHolder;
    @Autowired
    private CommentService commentService;
    @Autowired
    private EventProducer eventProducer;
    @Autowired
    private DiscussPostService discussPostService;
    @Autowired
    private RedisTemplate redisTemplate;

    @RequestMapping(path = "/add/{PostId}",method = RequestMethod.POST)
    public String addComments(@PathVariable("PostId") int postId, Comment comment){
        //1.
        comment.setUserId(hostHolder.getUser().getId());
        comment.setCreateTime(new Date());
        comment.setStatus(0);
        //2.
        commentService.addComment(comment);
        //3.send system　comment message
        Event event = new Event()
                .setTopic(KAFKA_TOPIC_COMMENT)
                .setUserId(hostHolder.getUser().getId())
                .setEntityType(comment.getEntityType())
                .setEntityId(comment.getEntityId())
                .setData("postId",postId);
        if(comment.getEntityType() == ENTITY_TYPE_POST){
            //the author of post
            event.setEntityUserId(discussPostService.findDiscussPostById(comment.getEntityId()).getUserId());
        }else if(comment.getEntityType() == ENTITY_TYPE_COMMENT){
            //the author of comment
            event.setEntityUserId(commentService.findCommentById(comment.getEntityId()).getUserId());
        }
        eventProducer.fireEvent(event);

  
}

其他两类业务代码类似.

1.5.4)展示某个用户的通知列表

@RequestMapping(path = "/notice/list",method = RequestMethod.GET)
    public String getNoticeList(Model model){
        User user = hostHolder.getUser();
        //comment
        Map<String,Object> map = new HashMap<>();
        Message leastNotice = messageService.findLeastNotice(KAFKA_TOPIC_COMMENT, user.getId());
        if(leastNotice!=null){
            map.put("leastNotice",leastNotice);
            HashMap content = JSONObject.parseObject(leastNotice.getContent(), HashMap.class);
            map.put("entityType",content.get("entityType"));
            map.put("entityId",content.get("entityId"));
            map.put("user",userService.findUserById((Integer) content.get("userId")));
            map.put("postId",content.get("postId"));

            int noticeCount = messageService.findNoticeCount(KAFKA_TOPIC_COMMENT, user.getId());
            int unReadNoticeCount = messageService.findUnReadNoticeCount(KAFKA_TOPIC_COMMENT, user.getId());
            map.put("noticeCount",noticeCount);
            map.put("unReadNoticeCount",unReadNoticeCount);
            model.addAttribute("comment",map);
        }


        //like
        map = new HashMap<>();
        leastNotice = messageService.findLeastNotice(KAFKA_TOPIC_LIKE, user.getId());
        if(leastNotice!=null){
            map.put("leastNotice",leastNotice);
            HashMap content = JSONObject.parseObject(leastNotice.getContent(), HashMap.class);
            map.put("entityType",content.get("entityType"));
            map.put("entityId",content.get("entityId"));
            map.put("user",userService.findUserById((Integer) content.get("userId")));
            map.put("postId",content.get("postId"));

            int noticeCount = messageService.findNoticeCount(KAFKA_TOPIC_LIKE, user.getId());
            int unReadNoticeCount = messageService.findUnReadNoticeCount(KAFKA_TOPIC_LIKE, user.getId());
            map.put("noticeCount",noticeCount);
            map.put("unReadNoticeCount",unReadNoticeCount);
            model.addAttribute("like",map);
        }


        //follow
        map = new HashMap<>();
        leastNotice = messageService.findLeastNotice(KAFKA_TOPIC_FOLLOW, user.getId());
        if(leastNotice!=null){
            map.put("leastNotice",leastNotice);
            HashMap content = JSONObject.parseObject(leastNotice.getContent(), HashMap.class);
            map.put("entityType",content.get("entityType"));
            map.put("entityId",content.get("entityId"));
            map.put("user",userService.findUserById((Integer) content.get("userId")));
            int noticeCount = messageService.findNoticeCount(KAFKA_TOPIC_FOLLOW, user.getId());
            int unReadNoticeCount = messageService.findUnReadNoticeCount(KAFKA_TOPIC_FOLLOW, user.getId());
            map.put("noticeCount",noticeCount);
            map.put("unReadNoticeCount",unReadNoticeCount);
            model.addAttribute("follow",map);
        }


        //
        model.addAttribute("unreadConversationCount",messageService.findUnreadLetterCount(user.getId(), null));
        model.addAttribute("unreadNoticeCount",messageService.findUnReadNoticeCount(null,user.getId()));

        return "/site/notice";
    }

1.5.5)某类通知的详情

…..

总结:到此就实现了在spring项目中用kafka作为消息引擎系统来实现系统通知的功能.