feat(server): 增加工作流循环调用检测

- 新增 TreeCycleDetectionWithCache 类用于检测工作流中的循环调用- 在 WorkflowServiceImpl 中调用循环检测逻辑，确保工作流中不存在循环调用
- 优化 WorkflowExecutorActor 和 WorkflowBatchHandler 中的任务批次处理逻辑

snail-job-server/snail-job-server-common/src/main/java/com/aizuda/snailjob/server/common/util/TreeCycleDetectionWithCache.java

@@ -0,0 +1,101 @@
+package com.aizuda.snailjob.server.common.util;
+
+import com.aizuda.snailjob.common.core.constant.SystemConstants;
+import com.aizuda.snailjob.common.core.context.SnailSpringContext;
+import com.aizuda.snailjob.common.core.util.JsonUtil;
+import com.aizuda.snailjob.server.common.dto.WorkflowConfig;
+import com.aizuda.snailjob.template.datasource.persistence.mapper.WorkflowNodeMapper;
+import com.aizuda.snailjob.template.datasource.persistence.po.WorkflowNode;
+import com.baomidou.mybatisplus.core.conditions.query.LambdaQueryWrapper;
+import lombok.Data;
+
+import java.util.*;
+
+/**
+ * 树形结构循环检测
+ * 检测工作流中调用工作流，其中是否存在循环调用
+ */
+public class TreeCycleDetectionWithCache {
+
+    @Data
+    public static class Node {
+
+        /**
+         * 工作流ID
+         */
+        long id;
+
+        public Node(long id) {
+            this.id = id;
+        }
+
+        /**
+         * 从数据库中该工作流中调用的所有子工作流
+         */
+        public List<Node> getChildrenFromDB() {
+            WorkflowNodeMapper workflowNodeMapper = SnailSpringContext.getBean(WorkflowNodeMapper.class);
+            assert workflowNodeMapper != null;
+            List<WorkflowNode> workflowNodes = workflowNodeMapper.selectList(new LambdaQueryWrapper<WorkflowNode>()
+                    .eq(WorkflowNode::getJobId, SystemConstants.WORKFLOW_JOB_ID)
+                    .eq(WorkflowNode::getWorkflowId, id));
+            ArrayList<Node> nodes = new ArrayList<>(workflowNodes.size());
+            for (WorkflowNode workflowNode : workflowNodes) {
+                WorkflowConfig nodeInfo = JsonUtil.parseObject(workflowNode.getNodeInfo(), WorkflowConfig.class);
+                nodes.add(new Node(nodeInfo.getId()));
+            }
+            return nodes;
+        }
+    }
+
+    /**
+     * 缓存子节点,防止多次查询数据库
+     */
+    private final Map<Node, List<Node>> childCache = new HashMap<>();
+    /**
+     * 缓存已访问的节点，防止重复处理
+     */
+    private final Set<Node> visited = new HashSet<>();
+    /**
+     * 递归栈，用于检测环
+     */
+    private final Set<Node> recursionStack = new HashSet<>();
+
+    /**
+     * 检测环
+     */
+    public boolean hasCycle(Node root) {
+        return dfs(root);
+    }
+
+    private boolean dfs(Node node) {
+        if (recursionStack.contains(node)) {
+            return true; // 发现环
+        }
+
+        if (visited.contains(node)) {
+            return false; // 已经处理过，无需重复处理
+        }
+
+        visited.add(node);
+        recursionStack.add(node);
+
+        // 获取子节点（可能来自缓存）
+        List<Node> children = getCachedChildren(node);
+        for (Node child : children) {
+            if (dfs(child)) {
+                return true;
+            }
+        }
+
+        recursionStack.remove(node);
+        return false;
+    }
+
+    private List<Node> getCachedChildren(Node node) {
+        if (!childCache.containsKey(node)) {
+            childCache.put(node, node.getChildrenFromDB());
+        }
+        return childCache.get(node);
+    }
+
+}

snail-job-server/snail-job-server-job-task/src/main/java/com/aizuda/snailjob/server/job/task/support/dispatch/WorkflowExecutorActor.java

@@ -101,7 +101,7 @@ public class WorkflowExecutorActor extends AbstractActor {
         if (SystemConstants.ROOT.equals(taskExecute.getParentId())
                 && JobTaskBatchStatusEnum.WAITING.getStatus() == workflowTaskBatch.getTaskBatchStatus()) {
             handlerTaskBatch(taskExecute, JobTaskBatchStatusEnum.RUNNING.getStatus(),
-                    JobOperationReasonEnum.NONE.getReason());
+                    null);
 
             Workflow workflow = workflowMapper.selectById(workflowTaskBatch.getWorkflowId());
             JobTimerWheel.clearCache(
@@ -325,7 +325,7 @@ public class WorkflowExecutorActor extends AbstractActor {
      * @param taskStatus      任务批次状态
      * @param operationReason 操作原因
      */
-    private void handlerTaskBatch(WorkflowNodeTaskExecuteDTO taskExecute, int taskStatus, int operationReason) {
+    private void handlerTaskBatch(WorkflowNodeTaskExecuteDTO taskExecute, int taskStatus, Integer operationReason) {
 
         WorkflowTaskBatch jobTaskBatch = new WorkflowTaskBatch();
         jobTaskBatch.setId(taskExecute.getWorkflowTaskBatchId());

snail-job-server/snail-job-server-job-task/src/main/java/com/aizuda/snailjob/server/job/task/support/handler/WorkflowBatchHandler.java

@@ -137,7 +137,7 @@ public class WorkflowBatchHandler {
 
         // 判定最后的工作流批次状态
         int taskStatus = JobTaskBatchStatusEnum.SUCCESS.getStatus();
-        int operationReason = JobOperationReasonEnum.NONE.getReason();
+//        int operationReason = JobOperationReasonEnum.NONE.getReason();
 
         // 判定所有的叶子节点是否完成
         List<Long> leaves = MutableGraphCache.getLeaves(workflowTaskBatchId, flowInfo);
@@ -175,7 +175,7 @@ public class WorkflowBatchHandler {
             }
         }
 
-        handlerTaskBatch(workflowTaskBatchId, taskStatus, operationReason);
+        handlerTaskBatch(workflowTaskBatchId, taskStatus, null);
 
         return true;
 
@@ -184,12 +184,12 @@ public class WorkflowBatchHandler {
     /**
      * 修改工作流任务批次状态
      */
-    private void handlerTaskBatch(Long workflowTaskBatchId, int taskStatus, int operationReason) {
+    private void handlerTaskBatch(Long workflowTaskBatchId, int taskStatus, Integer operationReason) {
 
         WorkflowTaskBatch jobTaskBatch = new WorkflowTaskBatch();
         jobTaskBatch.setId(workflowTaskBatchId);
         jobTaskBatch.setTaskBatchStatus(taskStatus);
-//        jobTaskBatch.setOperationReason(operationReason);
+        jobTaskBatch.setOperationReason(operationReason);
         workflowTaskBatchMapper.updateById(jobTaskBatch);
     }
 

snail-job-server/snail-job-server-web/src/main/java/com/aizuda/snailjob/server/web/service/impl/WorkflowServiceImpl.java

@@ -246,6 +246,12 @@ public class WorkflowServiceImpl implements WorkflowService {
 
         log.info("Graph construction complete. graph:[{}]", graph);
 
+        TreeCycleDetectionWithCache detector = new TreeCycleDetectionWithCache();
+        TreeCycleDetectionWithCache.Node root = new TreeCycleDetectionWithCache.Node(workflowRequestVO.getId());
+        //是否存在循环
+        boolean hasCycle = detector.hasCycle(root);
+        Assert.isFalse(hasCycle, () -> new SnailJobServerException("工作流中存在循环调用"));
+
         // 保存图信息
         workflow = WorkflowConverter.INSTANCE.convert(workflowRequestVO);
         workflow.setId(workflowRequestVO.getId());