update readme

CLAUDE.md

@@ -72,12 +72,12 @@ The workflow engine is organized into 8 layers:
    - `IFlowNode` - Interface defining node lifecycle methods
    - `BaseFlowNode` - Abstract base for all nodes, manages actions and strategies
    - `BaseAuditNode` - Abstract base for audit nodes (ApprovalNode, HandleNode)
-   - 11 node types: StartNode, EndNode, ApprovalNode, HandleNode, NotifyNode, ConditionBranchNode, ParallelBranchNode, RouterBranchNode, InclusiveBranchNode, DelayNode, TriggerNode, SubProcessNode
+   - 12 node types: StartNode, EndNode, ApprovalNode, HandleNode, NotifyNode, BranchNodeBranchNode, ParallelBranchNode, RouterBranchNode, InclusiveBranchNode, SubProcessNode, DelayNode, TriggerNode
 
 3. **Action Layer** (`com.codingapi.flow.action`, `com.codingapi.flow.action.actions`)
    - `IFlowAction` - Interface for node actions with `copy()` method
    - `BaseAction` - Abstract base with `triggerNode()` for recursive traversal
-   - 8 action types: DefaultAction, PassAction, RejectAction, SaveAction, ReturnAction, TransferAction, AddAuditAction, DelegateAction, CustomAction
+   - 9 action types: DefaultAction, PassAction, RejectAction, SaveAction, ReturnAction, TransferAction, AddAuditAction, DelegateAction, CustomAction
 
 4. **Record Layer** (`com.codingapi.flow.record`)
    - `FlowRecord` - Execution record with states (TODO/DONE, RUNNING/FINISH/ERROR/DELETE)
@@ -100,6 +100,13 @@ The workflow engine is organized into 8 layers:
    - `ScriptRuntimeContext` - Groovy script execution with thread-safe design and auto-cleanup
    - Script types: OperatorMatchScript, OperatorLoadScript, NodeTitleScript, ConditionScript, ErrorTriggerScript, RejectActionScript
 
+### Supporting Architectures
+
+- **Repository Pattern** (`com.codingapi.flow.repository`) - Abstraction for data persistence, isolates framework from implementation
+- **Gateway Pattern** (`com.codingapi.flow.gateway`) - Anti-corruption layer for external system integration (operators, users)
+- **Backup System** (`com.codingapi.flow.backup`) - Workflow versioning and backup management
+- **Error Handling** (`com.codingapi.flow.error`) - Centralized error throwing mechanism for flow control
+
 ### Node Lifecycle (Critical for Understanding Flow)
 
 When a node is executed, methods are called in this order:
@@ -139,6 +146,8 @@ When encountering `ParallelBranchNode`:
 - **Chain of Responsibility**: `triggerNode()` recursive traversal, `StrategyManager` strategy iteration
 - **Composite**: Nodes contain multiple strategies and actions
 - **Copy Pattern**: `INodeStrategy.copy()`, `IFlowAction.copy()`, `BaseFlowNode.setActions()`, `BaseFlowNode.setStrategies()`
+- **Repository**: Abstracts data persistence, implementations in infra layer
+- **Gateway**: Anti-corruption layer for external system integration
 
 ### Strategy-Driven Node Configuration (Critical Architecture)
 
@@ -200,6 +209,8 @@ Use framework exceptions instead of generic exceptions (IllegalArgumentException
 - **Custom Strategies**: Extend `BaseStrategy`, implement `INodeStrategy`
 - **Custom Scripts**: Use `ScriptRuntimeContext` for Groovy execution
 - **Event Listeners**: Listen to `FlowRecordStartEvent`, `FlowRecordTodoEvent`, `FlowRecordDoneEvent`, `FlowRecordFinishEvent`
+- **Repository Implementations**: Implement repository interfaces in infra layer for persistence
+- **Gateway Implementations**: Implement gateway interfaces for system integration
 
 ## Documentation References
 

Design.md

@@ -17,13 +17,16 @@
     - `operatorCreateScript`: 创建者匹配脚本
     - `isInterfere`: 是否开启干预
     - `isRevoke`: 是否开启撤销
+    - `createdOperator`: 创建者ID
+    - `createdTime`: 创建时间
+    - `schema`: 流程版本标识
 - **核心方法**:
     - `verify()`: 验证流程定义的合法性
     - `nextNodes(IFlowNode)`: 获取指定节点的后续节点
     - `getStartNode()`: 获取开始节点
     - `getEndNode()`: 获取结束节点
     - `toJson()`: 序列化为JSON
-    - `formJson()`: 从JSON反序列化
+    - `formJson(String)`: 从JSON反序列化（静态方法）
 
 #### WorkflowBuilder
 - **位置**: `com.codingapi.flow.workflow.WorkflowBuilder`
@@ -70,7 +73,7 @@
     - `isDone()`: 通过StrategyManager判断多人审批完成状态
     - `generateCurrentRecords()`: 通过StrategyManager加载操作者并生成记录
 
-#### 节点类型一览 (11种)
+#### 节点类型一览 (12种)
 
 | 节点类型 | 类名 | NODE_TYPE | 说明 |
 |---------|------|-----------|------|
@@ -167,7 +170,12 @@
 #### FlowAdvice
 - **位置**: `com.codingapi.flow.session.FlowAdvice`
 - **职责**: 封装审批操作的相关参数
-- **核心属性**: `advice`(审批意见), `signKey`(签名), `action`(动作), `backNode`(退回节点), `transferOperators`(转办人员)
+- **核心属性**:
+    - `advice`: 审批意见
+    - `signKey`: 签名
+    - `action`: 动作类型
+    - `backNode`: 退回节点（类型为 `IFlowNode`，节点对象引用）
+    - `transferOperators`: 转办人员列表
 
 ---
 

README.md

@@ -25,20 +25,37 @@ flow-engine
 │   ├── src/main/java
 │   │   └── com.codingapi.flow
 │   │       ├── action            # 操作层
+│   │       │   ├── actions       # 具体操作实现
+│   │       │   └── factory       # 操作工厂
+│   │       ├── backup            # 流程备份
 │   │       ├── builder           # 构建器
 │   │       ├── context           # 上下文
+│   │       ├── edge              # 节点连线
+│   │       ├── error             # 错误处理
 │   │       ├── event             # 事件
-│   │       ├── exception         # 异常
-│   │       ├── form              # 表单
-│   │       ├── gateway           # 网关
+│   │       ├── exception         # 异常体系
+│   │       ├── form              # 表单系统
+│   │       │   └── permission    # 字段权限
+│   │       ├── gateway           # 网关接口防腐层
 │   │       ├── node              # 节点层
-│   │       ├── operator          # 操作人
+│   │       │   ├── factory       # 节点工厂
+│   │       │   ├── helper        # 节点辅助类
+│   │       │   ├── manager       # 节点管理器
+│   │       │   └── nodes         # 具体节点实现
+│   │       ├── operator          # 操作人接口
 │   │       ├── pojo              # 数据对象
-│   │       ├── record            # 记录层
-│   │       ├── script            # 脚本层
+│   │       │   ├── body          # 请求体
+│   │       │   └── request       # 请求对象
+│   │       ├── record            # 流程记录
+│   │       ├── repository        # 仓储接口
+│   │       ├── script            # 脚本系统
+│   │       │   ├── action        # 操作脚本
+│   │       │   ├── node          # 节点脚本
+│   │       │   └── runtime       # 脚本运行时
+│   │       ├── service           # 服务层
+│   │       │   └── impl          # 服务实现
 │   │       ├── session           # 会话层
 │   │       ├── strategy          # 策略层
-│   │       ├── user              # 用户
 │   │       ├── utils             # 工具类
 │   │       └── workflow          # 流程层
 │   └── src/test/java             # 测试代码

flow-engine-framework/src/main/java/com/codingapi/flow/context/RepositoryContext.java

@@ -60,4 +60,8 @@ public int getParallelBranchTriggerCount(String parallelId) {
     public void addParallelTriggerCount(String parallelId) {
         parallelBranchRepository.addTriggerCount(parallelId);
     }
+
+    public void clearParallelTriggerCount(String parallelId) {
+        parallelBranchRepository.clearTriggerCount(parallelId);
+    }
 }

flow-engine-framework/src/main/java/com/codingapi/flow/node/BaseAuditNode.java

@@ -107,7 +107,8 @@ public boolean isDone(FlowSession session) {
     @Override
     public List<FlowRecord> generateCurrentRecords(FlowSession session) {
 
-        if (this.isWaitParallelRecord(session)) {
+        // 是否等待并行合并节点
+        if (this.isWaitRecordMargeParallelNode(session)) {
             return List.of();
         }
 

flow-engine-framework/src/main/java/com/codingapi/flow/node/BaseFlowNode.java

@@ -171,7 +171,7 @@ private void verifyDefaultConfig(){
     /**
      * 是否等待并行节点的汇聚
      */
-    public boolean isWaitParallelRecord(FlowSession session) {
+    public boolean isWaitRecordMargeParallelNode(FlowSession session) {
         FlowRecord currentRecord = session.getCurrentRecord();
         if (currentRecord != null && this.getId().equals(currentRecord.getParallelBranchNodeId())) {
             RepositoryContext.getInstance().addParallelTriggerCount(currentRecord.getParallelId());
@@ -180,6 +180,7 @@ public boolean isWaitParallelRecord(FlowSession session) {
             if (parallelBranchCount == parallelBranchTotal) {
                 // 清空并行节点，防止数据继续继承到后续节点
                 currentRecord.clearParallel();
+                RepositoryContext.getInstance().clearParallelTriggerCount(currentRecord.getParallelId());
             }
             return parallelBranchCount != parallelBranchTotal;
         }

flow-engine-framework/src/main/java/com/codingapi/flow/node/nodes/EndNode.java

@@ -30,7 +30,7 @@ public String getType() {
 
     @Override
     public List<FlowRecord> generateCurrentRecords(FlowSession session) {
-        if(this.isWaitParallelRecord(session)){
+        if(this.isWaitRecordMargeParallelNode(session)){
             return List.of();
         }
         // 构建结束记录

flow-engine-framework/src/main/java/com/codingapi/flow/repository/ParallelBranchRepository.java

@@ -5,4 +5,7 @@ public interface ParallelBranchRepository {
     int getTriggerCount(String parallelId);
 
     void addTriggerCount(String parallelId);
+
+    void clearTriggerCount(String parallelId);
+
 }

flow-engine-framework/src/test/java/com/codingapi/flow/repository/ParallelBranchRepositoryImpl.java

@@ -17,4 +17,9 @@ public int getTriggerCount(String parallelId) {
     public void addTriggerCount(String parallelId) {
         this.cache.put(parallelId, this.getTriggerCount(parallelId) + 1);
     }
+
+    @Override
+    public void clearTriggerCount(String parallelId) {
+        this.cache.remove(parallelId);
+    }
 }