v2.8.1_e初始化

1、测试代码，开源用户尽量不要使用；
2、包含Kafka-HA的相关功能，在v2.8.0_e的基础上，补充按照clientId切换的功能；
3、基于v2.8.0_e拉的分支；

kafka-manager-task/src/main/java/com/xiaojukeji/kafka/manager/task/dispatch/ha/HaFlushASSwitchJob.java

@@ -1,4 +1,4 @@
-package com.xiaojukeji.kafka.manager.task.dispatch.op;
+package com.xiaojukeji.kafka.manager.task.dispatch.ha;
 
 import com.xiaojukeji.kafka.manager.service.biz.job.HaASSwitchJobManager;
 import com.xiaojukeji.kafka.manager.service.service.ha.HaASSwitchJobService;

kafka-manager-task/src/main/java/com/xiaojukeji/kafka/manager/task/dispatch/ha/HandleHaClientNewTopic.java

@@ -0,0 +1,156 @@
+//package com.xiaojukeji.kafka.manager.task.dispatch.ha;
+//
+//import com.xiaojukeji.kafka.manager.common.bizenum.ha.HaResTypeEnum;
+//import com.xiaojukeji.kafka.manager.common.bizenum.ha.job.HaJobStatusEnum;
+//import com.xiaojukeji.kafka.manager.common.constant.ConfigConstant;
+//import com.xiaojukeji.kafka.manager.common.constant.Constant;
+//import com.xiaojukeji.kafka.manager.common.entity.dto.ha.ASSwitchJobDTO;
+//import com.xiaojukeji.kafka.manager.common.entity.pojo.ClusterDO;
+//import com.xiaojukeji.kafka.manager.common.entity.pojo.ha.HaASRelationDO;
+//import com.xiaojukeji.kafka.manager.common.entity.pojo.ha.HaASSwitchJobDO;
+//import com.xiaojukeji.kafka.manager.common.utils.HAUtils;
+//import com.xiaojukeji.kafka.manager.common.utils.Tuple;
+//import com.xiaojukeji.kafka.manager.common.utils.ValidateUtils;
+//import com.xiaojukeji.kafka.manager.service.biz.job.HaASSwitchJobManager;
+//import com.xiaojukeji.kafka.manager.service.service.ClusterService;
+//import com.xiaojukeji.kafka.manager.service.service.ConfigService;
+//import com.xiaojukeji.kafka.manager.service.service.gateway.TopicConnectionService;
+//import com.xiaojukeji.kafka.manager.service.service.ha.HaASRelationService;
+//import com.xiaojukeji.kafka.manager.service.service.ha.HaASSwitchJobService;
+//import com.xiaojukeji.kafka.manager.task.component.AbstractScheduledTask;
+//import com.xiaojukeji.kafka.manager.task.component.CustomScheduled;
+//import org.springframework.beans.factory.annotation.Autowired;
+//import org.springframework.stereotype.Component;
+//
+//import java.util.*;
+//import java.util.function.Function;
+//import java.util.stream.Collectors;
+//
+///**
+// * 主备切换任务
+// */
+//@Component
+//@CustomScheduled(name = "HandleHaClientNewTopic",
+//        cron = "0 0/2 * * * ?",
+//        threadNum = 1,
+//        description = "处理HAClient的新增Topic")
+//public class HandleHaClientNewTopic extends AbstractScheduledTask<ClusterDO> {
+//    @Autowired
+//    private ClusterService clusterService;
+//
+//    @Autowired
+//    private HaASRelationService haASRelationService;
+//
+//    @Autowired
+//    private TopicConnectionService topicConnectionService;
+//
+//    @Autowired
+//    private HaASSwitchJobManager haASSwitchJobManager;
+//
+//    @Autowired
+//    private HaASSwitchJobService haASSwitchJobService;
+//
+//    @Autowired
+//    private ConfigService configService;
+//
+//    @Override
+//    public List<ClusterDO> listAllTasks() {
+//        return clusterService.list();
+//    }
+//
+//    @Override
+//    public void processTask(ClusterDO clusterDO) {
+//        if (this.existRunningTask(clusterDO.getId())) {
+//            // 存在运行中的任务
+//            return;
+//        }
+//
+//        // 获取已经建立HA的Client
+//        List<HaASRelationDO> doList = haASRelationService.listAllHAFromDB(clusterDO.getId(), HaResTypeEnum.KAFKA_USER_AND_CLIENT);
+//
+//        // 获取已经建立HA的Topic
+//        Map<String, HaASRelationDO> nameMap = haASRelationService.listAllHAFromDB(clusterDO.getId(), HaResTypeEnum.TOPIC)
+//                .stream()
+//                .collect(Collectors.toMap(HaASRelationDO::getActiveResName, Function.identity()));
+//
+//        // 新的主备集群 & 需要切换的Topic
+//        Long newActiveClusterId = null;
+//        Long newStandbyClusterId = null;
+//        Map<String, HaASRelationDO> needSwitchTopicMap = new HashMap<>();
+//
+//        // 查找clientId关联的Topic列表
+//        for (HaASRelationDO asRelationDO: doList) {
+//            if (newActiveClusterId != null && !newActiveClusterId.equals(asRelationDO.getActiveClusterPhyId())) {
+//                // 一次切换，仅能有一个主集群ID，不能有多个。不一致时，直接忽略
+//                continue;
+//            }
+//
+//            Tuple<String, String> userAndClient = HAUtils.splitKafkaUserAndClient(asRelationDO.getActiveResName());
+//            if (userAndClient == null || ValidateUtils.isBlank(userAndClient.getV2())) {
+//                continue;
+//            }
+//
+//            // 获取该client对应的Topic
+//            Set<String> topicNameSet = topicConnectionService.getKafkaUserAndClientIdTopicNames(
+//                    new HashSet<>(Arrays.asList(asRelationDO.getActiveClusterPhyId(), asRelationDO.getStandbyClusterPhyId())),
+//                    userAndClient.getV1(),
+//                    userAndClient.getV2(),
+//                    new Date(System.currentTimeMillis() - configService.getLongValue(ConfigConstant.HA_CONNECTION_ACTIVE_TIME_UNIT_MIN, 20L) * 60L * 1000L),
+//                    new Date()
+//            );
+//
+//            // 遍历Topic，判断主备关系是否符合预期
+//            for (String topicName: topicNameSet) {
+//                HaASRelationDO topicRelation = nameMap.get(topicName);
+//                if (topicRelation == null
+//                        || asRelationDO.getActiveClusterPhyId().equals(topicRelation.getActiveClusterPhyId())) {
+//                    // Topic为空，未建立高可用，忽略该Topic
+//                    // 已建立HA，且该Topic的主备信息和当前clientId一致，因此也不需要进行主备切换
+//                    continue;
+//                }
+//
+//                // 主备信息不一致时，进行主备切换
+//                if (needSwitchTopicMap.isEmpty()) {
+//                    newActiveClusterId = asRelationDO.getActiveClusterPhyId();
+//                    newStandbyClusterId = asRelationDO.getStandbyClusterPhyId();
+//                }
+//
+//                needSwitchTopicMap.put(topicName, topicRelation);
+//            }
+//        }
+//
+//        if (this.existRunningTask(clusterDO.getId())) {
+//            // 再次判断是否存在运行中的任务
+//            return;
+//        }
+//
+//        // 创建任务
+//        haASSwitchJobManager.createJob(
+//                this.convert2ASSwitchJobDTO(newActiveClusterId, newStandbyClusterId, new ArrayList<>(needSwitchTopicMap.values())),
+//                Constant.DEFAULT_USER_NAME
+//        );
+//    }
+//
+//    private ASSwitchJobDTO convert2ASSwitchJobDTO(Long newActiveClusterId, Long newStandbyClusterId, List<HaASRelationDO> doList) {
+//        ASSwitchJobDTO dto = new ASSwitchJobDTO();
+//        dto.setAll(false);
+//        dto.setMustContainAllKafkaUserTopics(false);
+//        dto.setActiveClusterPhyId(newActiveClusterId);
+//        dto.setStandbyClusterPhyId(newStandbyClusterId);
+//        dto.setTopicNameList(doList.stream().map(elem -> elem.getActiveResName()).collect(Collectors.toList()));
+//        dto.setKafkaUserAndClientIdList(new ArrayList<>()); // clientId 或者 kafkaUser 已切换好，所以后台任务不需要执行该步骤
+//
+//        return dto;
+//    }
+//
+//    private boolean existRunningTask(Long clusterPhyId) {
+//        Map<Long/*集群ID*/, HaASSwitchJobDO> jobMap = haASSwitchJobService.listClusterLatestJobs();
+//
+//        HaASSwitchJobDO jobDO = jobMap.remove(clusterPhyId);
+//        if (jobDO == null || !HaJobStatusEnum.isRunning(jobDO.getJobStatus())) {
+//            return false;
+//        }
+//
+//        return true;
+//    }
+//}