@lobehub/lobehub 2.0.0-next.32 → 2.0.0-next.33

package/src/store/chat/slices/aiChat/actions/__tests__/generateAIChatV2.test.ts

@@ -131,7 +131,7 @@ describe('generateAIChatV2 actions', () => {
           await result.current.sendMessage({ message: TEST_CONTENT.USER_MESSAGE });
         });
 
-        expect(messageService.createNewMessage).not.toHaveBeenCalled();
+        expect(messageService.createMessage).not.toHaveBeenCalled();
         expect(result.current.internal_execAgentRuntime).not.toHaveBeenCalled();
       });
 
@@ -142,7 +142,7 @@ describe('generateAIChatV2 actions', () => {
           await result.current.sendMessage({ message: TEST_CONTENT.EMPTY });
         });
 
-        expect(messageService.createNewMessage).not.toHaveBeenCalled();
+        expect(messageService.createMessage).not.toHaveBeenCalled();
       });
 
       it('should not send when message is empty with empty files array', async () => {
@@ -152,7 +152,7 @@ describe('generateAIChatV2 actions', () => {
           await result.current.sendMessage({ message: TEST_CONTENT.EMPTY, files: [] });
         });
 
-        expect(messageService.createNewMessage).not.toHaveBeenCalled();
+        expect(messageService.createMessage).not.toHaveBeenCalled();
       });
     });
 
@@ -312,7 +312,7 @@ describe('generateAIChatV2 actions', () => {
           });
         });
 
-        expect(messageService.createNewMessage).toHaveBeenCalled();
+        expect(messageService.createMessage).toHaveBeenCalled();
         expect(result.current.internal_execAgentRuntime).not.toHaveBeenCalled();
       });
 
@@ -749,7 +749,7 @@ describe('generateAIChatV2 actions', () => {
 
     beforeEach(() => {
       // Reset mocks
-      vi.spyOn(messageService, 'createNewMessage').mockResolvedValue({
+      vi.spyOn(messageService, 'createMessage').mockResolvedValue({
         id: 'new-assistant-block-id',
         messages: [] as any,
       });
@@ -814,8 +814,8 @@ describe('generateAIChatV2 actions', () => {
         }),
       );
 
-      // Verify that createNewMessage was called with message params
-      expect(messageService.createNewMessage).toHaveBeenCalledWith(
+      // Verify that createMessage was called with message params
+      expect(messageService.createMessage).toHaveBeenCalledWith(
         expect.objectContaining({
           role: 'assistant',
           parentId: TOOL_RESULT_MSG_ID,
@@ -866,7 +866,7 @@ describe('generateAIChatV2 actions', () => {
         }),
       );
 
-      expect(messageService.createNewMessage).toHaveBeenCalledWith(
+      expect(messageService.createMessage).toHaveBeenCalledWith(
         expect.objectContaining({
           role: 'assistant',
           parentId: 'non-existent-tool-result-id',

package/src/store/chat/slices/aiChat/actions/__tests__/helpers.ts

@@ -58,7 +58,7 @@ export const createMockAbortController = () => {
  */
 export const spyOnMessageService = () => {
   const createMessageSpy = vi
-    .spyOn(messageService, 'createNewMessage')
+    .spyOn(messageService, 'createMessage')
     .mockResolvedValue({ id: TEST_IDS.NEW_MESSAGE_ID, messages: [] });
   const updateMessageSpy = vi
     .spyOn(messageService, 'updateMessage')

package/src/store/chat/slices/aiChat/actions/generateAIChatV2.ts

@@ -646,7 +646,7 @@ export const generateAIChatV2: StateCreator<
           payload.type,
         );
 
-        // 2. 使用 createNewMessage 创建 tool 消息
+        // 2. 使用 createMessage 创建 tool 消息
         const toolMessage: CreateNewMessageParams = {
           content: '',
           parentId: assistantId,

package/src/store/chat/slices/message/action.test.ts

@@ -25,7 +25,7 @@ vi.mock('@/services/message', () => ({
     removeMessage: vi.fn(),
     removeMessagesByAssistant: vi.fn(),
     removeMessages: vi.fn(() => Promise.resolve()),
-    createNewMessage: vi.fn(() => Promise.resolve({ id: 'new-message-id', messages: [] })),
+    createMessage: vi.fn(() => Promise.resolve({ id: 'new-message-id', messages: [] })),
     updateMessage: vi.fn(),
     removeAllMessages: vi.fn(() => Promise.resolve()),
   },
@@ -71,7 +71,7 @@ describe('chatMessage actions', () => {
         await result.current.addAIMessage();
       });
 
-      expect(messageService.createNewMessage).not.toHaveBeenCalled();
+      expect(messageService.createMessage).not.toHaveBeenCalled();
       expect(updateInputMessageSpy).not.toHaveBeenCalled();
     });
 
@@ -84,7 +84,7 @@ describe('chatMessage actions', () => {
         await result.current.addAIMessage();
       });
 
-      expect(messageService.createNewMessage).toHaveBeenCalledWith({
+      expect(messageService.createMessage).toHaveBeenCalledWith({
         content: inputMessage,
         role: 'assistant',
         sessionId: mockState.activeId,
@@ -113,7 +113,7 @@ describe('chatMessage actions', () => {
         await result.current.addUserMessage({ message: 'test message' });
       });
 
-      expect(messageService.createNewMessage).not.toHaveBeenCalled();
+      expect(messageService.createMessage).not.toHaveBeenCalled();
       expect(updateInputMessageSpy).not.toHaveBeenCalled();
     });
 
@@ -130,7 +130,7 @@ describe('chatMessage actions', () => {
         await result.current.addUserMessage({ message, fileList });
       });
 
-      expect(messageService.createNewMessage).toHaveBeenCalledWith({
+      expect(messageService.createMessage).toHaveBeenCalledWith({
         content: message,
         files: fileList,
         role: 'user',
@@ -154,7 +154,7 @@ describe('chatMessage actions', () => {
         await result.current.addUserMessage({ message });
       });
 
-      expect(messageService.createNewMessage).toHaveBeenCalledWith({
+      expect(messageService.createMessage).toHaveBeenCalledWith({
         content: message,
         files: undefined,
         role: 'user',
@@ -184,7 +184,7 @@ describe('chatMessage actions', () => {
         await result.current.addUserMessage({ message });
       });
 
-      expect(messageService.createNewMessage).toHaveBeenCalledWith({
+      expect(messageService.createMessage).toHaveBeenCalledWith({
         content: message,
         files: undefined,
         role: 'user',

package/src/store/chat/slices/message/action.ts

@@ -519,10 +519,10 @@ export const chatMessage: StateCreator<
     }
 
     try {
-      const result = await messageService.createNewMessage(message);
+      const result = await messageService.createMessage(message);
 
       if (!context?.skipRefresh) {
-        // Use the messages returned from createNewMessage (already grouped)
+        // Use the messages returned from createMessage (already grouped)
         replaceMessages(result.messages);
       }
 

package/src/store/chat/slices/plugin/action.test.ts

@@ -26,7 +26,7 @@ vi.mock('@/services/message', () => ({
     updateMessageError: vi.fn(),
     updateMessagePluginState: vi.fn(),
     updateMessagePluginArguments: vi.fn(),
-    createNewMessage: vi.fn(),
+    createMessage: vi.fn(),
   },
 }));
 
@@ -565,8 +565,8 @@ describe('ChatPluginAction', () => {
   describe('createAssistantMessageByPlugin', () => {
     it('should create an assistant message and replace messages', async () => {
       const mockMessages = [{ id: 'msg-1', content: 'test' }] as any;
-      // 模拟 messageService.createNewMessage 方法的实现
-      (messageService.createNewMessage as Mock).mockResolvedValue({
+      // 模拟 messageService.createMessage 方法的实现
+      (messageService.createMessage as Mock).mockResolvedValue({
         id: 'new-message-id',
         messages: mockMessages,
       });
@@ -588,8 +588,8 @@ describe('ChatPluginAction', () => {
         await result.current.createAssistantMessageByPlugin(content, parentId);
       });
 
-      // 验证 messageService.createNewMessage 是否被带有正确参数调用
-      expect(messageService.createNewMessage).toHaveBeenCalledWith({
+      // 验证 messageService.createMessage 是否被带有正确参数调用
+      expect(messageService.createMessage).toHaveBeenCalledWith({
         content,
         parentId,
         role: 'assistant',
@@ -604,7 +604,7 @@ describe('ChatPluginAction', () => {
     it('should handle errors when message creation fails', async () => {
       // 模拟 messageService.create 方法，使其抛出错误
       const errorMessage = 'Failed to create message';
-      (messageService.createNewMessage as Mock).mockRejectedValue(new Error(errorMessage));
+      (messageService.createMessage as Mock).mockRejectedValue(new Error(errorMessage));
 
       // 设置初始状态并模拟 refreshMessages 方法
       const initialState = {
@@ -626,7 +626,7 @@ describe('ChatPluginAction', () => {
       });
 
       // 验证 messageService.create 是否被带有正确参数调用
-      expect(messageService.createNewMessage).toHaveBeenCalledWith({
+      expect(messageService.createMessage).toHaveBeenCalledWith({
         content,
         parentId,
         role: 'assistant',

package/src/store/chat/slices/plugin/action.ts

@@ -98,7 +98,7 @@ export const chatPlugin: StateCreator<
       topicId: get().activeTopicId, // if there is activeTopicId，then add it to topicId
     };
 
-    const result = await messageService.createNewMessage(newMessage);
+    const result = await messageService.createMessage(newMessage);
     get().replaceMessages(result.messages);
   },
 

package/packages/context-engine/ARCHITECTURE.md

@@ -1,425 +0,0 @@
-# Context Engine 架构设计
-
-## 概述
-
-Context Engine 是一个灵活的消息处理管道系统，用于在 AI 对话中动态管理和处理上下文信息。它提供了一种可扩展的方式来注入、转换和验证消息流。
-
-## 核心概念
-
-### 1. 处理器（Processor）
-
-处理器是 Context Engine 的基本单元，负责对消息进行特定的处理操作。
-
-#### 处理器分类
-
-根据功能职责，处理器分为以下几类：
-
-1. **注入器（Injector）**
-   - 职责：向消息流中添加新的上下文信息
-   - 示例：SystemRoleInjector、HistoryInjector、RAGContextInjector
-
-2. **转换器（Transformer）**
-   - 职责：修改现有消息的内容或结构
-   - 示例：MessageRoleTransformer、ImageContentProcessor
-
-3. **截断器（Truncator）**
-   - 职责：根据特定规则裁剪消息内容
-   - 示例：HistoryTruncator、TokenBasedTruncator
-
-4. **验证器（Validator）**
-   - 职责：验证消息是否符合特定要求
-   - 示例：ModelCapabilityValidator
-
-5. **重排器（Reorderer）**
-   - 职责：调整消息的顺序
-   - 示例：ToolMessageReorder
-
-### 2. 管道（Pipeline）
-
-管道是一系列处理器的有序组合，消息流经管道时会被逐个处理器处理。
-
-```typescript
-interface Pipeline {
-  // 添加处理器到管道
-  add(processor: BaseProcessor): Pipeline;
-  
-  // 执行管道处理
-  execute(context: ProcessorContext): Promise<ProcessorContext>;
-  
-  // 获取管道中的所有处理器
-  getProcessors(): BaseProcessor[];
-}
-```
-
-### 3. 上下文（Context）
-
-上下文包含了处理器需要的所有信息：
-
-```typescript
-interface ProcessorContext {
-  messages: Message[];           // 消息列表
-  metadata?: ProcessorMetadata;  // 元数据
-  variables?: Record<string, any>; // 变量
-  abortSignal?: AbortSignal;    // 中止信号
-}
-```
-
-## 架构设计
-
-### 类层次结构
-
-```
-BaseProcessor (抽象基类)
-├── BaseInjector (注入器基类)
-│   ├── SystemRoleInjector
-│   ├── HistoryInjector
-│   ├── RAGContextInjector
-│   └── ...
-├── BaseTransformer (转换器基类)
-│   ├── MessageRoleTransformer
-│   └── ImageContentProcessor
-├── BaseTruncator (截断器基类)
-│   ├── HistoryTruncator
-│   └── TokenBasedTruncator
-├── BaseValidator (验证器基类)
-│   └── ModelCapabilityValidator
-└── BaseReorderer (重排器基类)
-    └── ToolMessageReorder
-```
-
-### 核心接口设计
-
-#### BaseProcessor
-
-```typescript
-abstract class BaseProcessor {
-  // 处理器类型
-  abstract readonly type: ProcessorType;
-  
-  // 处理器名称
-  abstract readonly name: string;
-  
-  // 主处理方法
-  async process(context: ProcessorContext): Promise<ProcessorContext> {
-    // 1. 前置验证
-    this.validateInput(context);
-    
-    // 2. 执行处理
-    const result = await this.doProcess(context);
-    
-    // 3. 后置验证
-    this.validateOutput(result);
-    
-    return result;
-  }
-  
-  // 子类实现的核心处理逻辑
-  protected abstract doProcess(context: ProcessorContext): Promise<ProcessorContext>;
-  
-  // 输入验证（可选覆盖）
-  protected validateInput(context: ProcessorContext): void {}
-  
-  // 输出验证（可选覆盖）
-  protected validateOutput(context: ProcessorContext): void {}
-}
-```
-
-#### BaseInjector
-
-```typescript
-abstract class BaseInjector extends BaseProcessor {
-  readonly type = ProcessorType.Injector;
-  
-  protected async doProcess(context: ProcessorContext): Promise<ProcessorContext> {
-    // 1. 判断是否需要注入
-    if (!this.shouldInject(context)) {
-      return context;
-    }
-    
-    // 2. 构建注入内容
-    const content = await this.buildContent(context);
-    
-    // 3. 创建消息
-    const message = this.createMessage(content, context);
-    
-    // 4. 确定注入位置
-    const position = this.getInjectionPosition(context);
-    
-    // 5. 执行注入
-    return this.inject(context, message, position);
-  }
-  
-  // 子类需要实现的方法
-  protected abstract shouldInject(context: ProcessorContext): boolean;
-  protected abstract buildContent(context: ProcessorContext): Promise<string>;
-  
-  // 可选覆盖的方法
-  protected getInjectionPosition(context: ProcessorContext): number {
-    return 0; // 默认注入到开头
-  }
-  
-  protected createMessage(content: string, context: ProcessorContext): Message {
-    return {
-      role: 'system',
-      content,
-      metadata: { injectedBy: this.name }
-    };
-  }
-}
-```
-
-## 使用方式
-
-### 1. 创建自定义处理器
-
-```typescript
-// 创建一个自定义注入器
-class CustomContextInjector extends BaseInjector {
-  readonly name = 'custom-context-injector';
-  
-  protected shouldInject(context: ProcessorContext): boolean {
-    // 判断逻辑
-    return !context.messages.some(msg => 
-      msg.metadata?.injectedBy === this.name
-    );
-  }
-  
-  protected async buildContent(context: ProcessorContext): Promise<string> {
-    // 构建内容逻辑
-    return `Custom context: ${context.variables?.customValue}`;
-  }
-}
-```
-
-### 2. 构建处理管道
-
-```typescript
-// 使用工厂模式创建管道
-const pipeline = createPipeline()
-  .add(new SystemRoleInjector())
-  .add(new HistoryInjector({ maxMessages: 10 }))
-  .add(new CustomContextInjector())
-  .add(new MessageRoleTransformer())
-  .add(new TokenBasedTruncator({ maxTokens: 4000 }))
-  .add(new ModelCapabilityValidator());
-
-// 执行管道
-const result = await pipeline.execute({
-  messages: initialMessages,
-  variables: { customValue: 'test' },
-  metadata: { model: 'gpt-4' }
-});
-```
-
-### 3. 条件处理
-
-```typescript
-// 基于条件的管道构建
-const pipeline = createPipeline();
-
-if (config.enableRAG) {
-  pipeline.add(new RAGContextInjector());
-}
-
-if (config.enableSearch) {
-  pipeline.add(new SearchContextInjector());
-}
-
-// 始终添加的处理器
-pipeline
-  .add(new HistoryTruncator())
-  .add(new ModelCapabilityValidator());
-```
-
-### 4. 错误处理
-
-```typescript
-try {
-  const result = await pipeline.execute(context);
-} catch (error) {
-  if (error instanceof ProcessorError) {
-    console.error(`Processor ${error.processorName} failed:`, error.message);
-  } else if (error instanceof PipelineError) {
-    console.error('Pipeline execution failed:', error.message);
-  }
-}
-```
-
-## 配置管理
-
-### 处理器配置
-
-每个处理器可以接受特定的配置选项：
-
-```typescript
-interface ProcessorConfig {
-  // 通用配置
-  enabled?: boolean;
-  priority?: number;
-  
-  // 处理器特定配置
-  [key: string]: any;
-}
-
-// 示例：历史注入器配置
-interface HistoryInjectorConfig extends ProcessorConfig {
-  maxMessages?: number;
-  includeSystemMessages?: boolean;
-  preserveTools?: boolean;
-}
-```
-
-### 管道配置
-
-```typescript
-interface PipelineConfig {
-  processors: Array<{
-    type: string;
-    config?: ProcessorConfig;
-  }>;
-  
-  // 全局配置
-  abortOnError?: boolean;
-  timeout?: number;
-}
-
-// 从配置创建管道
-const pipeline = createPipelineFromConfig({
-  processors: [
-    { type: 'system-role', config: { enabled: true } },
-    { type: 'history', config: { maxMessages: 20 } },
-    { type: 'rag', config: { threshold: 0.7 } }
-  ],
-  abortOnError: false,
-  timeout: 30000
-});
-```
-
-## 最佳实践
-
-### 1. 单一职责原则
-每个处理器应该只负责一种特定的处理任务。
-
-### 2. 可配置性
-处理器应该通过配置参数来控制行为，而不是硬编码。
-
-### 3. 错误处理
-- 使用具体的错误类型
-- 提供有意义的错误信息
-- 考虑错误恢复策略
-
-### 4. 性能优化
-- 避免不必要的消息复制
-- 使用流式处理处理大量数据
-- 实现适当的缓存机制
-
-### 5. 测试
-- 为每个处理器编写单元测试
-- 测试处理器组合的集成测试
-- 边界条件和错误场景测试
-
-## 扩展机制
-
-### 1. 自定义处理器类型
-
-```typescript
-// 定义新的处理器类型
-enum CustomProcessorType {
-  Analyzer = 'analyzer',
-  Enhancer = 'enhancer'
-}
-
-// 创建对应的基类
-abstract class BaseAnalyzer extends BaseProcessor {
-  readonly type = CustomProcessorType.Analyzer;
-  
-  // 分析器特定的方法
-  protected abstract analyze(messages: Message[]): AnalysisResult;
-}
-```
-
-### 2. 处理器组合
-
-```typescript
-// 创建复合处理器
-class CompositeProcessor extends BaseProcessor {
-  constructor(private processors: BaseProcessor[]) {
-    super();
-  }
-  
-  protected async doProcess(context: ProcessorContext): Promise<ProcessorContext> {
-    let result = context;
-    for (const processor of this.processors) {
-      result = await processor.process(result);
-    }
-    return result;
-  }
-}
-```
-
-### 3. 插件系统
-
-```typescript
-interface ProcessorPlugin {
-  name: string;
-  version: string;
-  processors: ProcessorDefinition[];
-}
-
-// 注册插件
-registry.registerPlugin({
-  name: 'custom-processors',
-  version: '1.0.0',
-  processors: [
-    { type: 'custom-injector', factory: () => new CustomInjector() },
-    { type: 'custom-validator', factory: () => new CustomValidator() }
-  ]
-});
-```
-
-## 迁移指南
-
-### 从当前架构迁移
-
-1. **BaseProvider 迁移到 BaseInjector**
-   ```typescript
-   // 旧代码
-   class MyProvider extends BaseProvider {
-     doProcess(context) {
-       // 实现
-     }
-   }
-   
-   // 新代码
-   class MyInjector extends BaseInjector {
-     shouldInject(context) {
-       // 判断逻辑
-     }
-     
-     buildContent(context) {
-       // 构建内容
-     }
-   }
-   ```
-
-2. **处理器分类**
-   - 将现有处理器按功能分类
-   - 继承对应的基类
-   - 实现必要的抽象方法
-
-3. **配置迁移**
-   - 统一配置格式
-   - 支持向后兼容
-   - 提供迁移工具
-
-## 总结
-
-新的 Context Engine 架构提供了：
-
-1. **清晰的分层结构**：基于功能的处理器分类
-2. **灵活的扩展机制**：易于添加新的处理器类型
-3. **强大的组合能力**：通过管道组合实现复杂功能
-4. **完善的错误处理**：细粒度的错误类型和恢复策略
-5. **优秀的可测试性**：模块化设计便于单元测试
-
-这种设计使得 Context Engine 更加模块化、可维护和可扩展，能够更好地满足不同场景下的上下文处理需求。