feat(api): 添加使用模拟数据的功能

app/lib/hooks/usePromptEnhancer.ts

@@ -43,11 +43,21 @@ export function usePromptEnhancer() {
             break;
           }
 
-          _input += decoder.decode(value);
+          const chunk = decoder.decode(value);
+          const lines = chunk.split('\n');
 
-          logger.trace('Set input', _input);
+          for (const line of lines) {
+            if (line.startsWith('0:')) {
+              const content = JSON.parse(line.slice(2));
+              _input += content;
+              setInput(_input);
+            } else if (line.startsWith('e:') || line.startsWith('d:')) {
+              // 处理元数据,如果需要的话
+              logger.trace('Metadata', line);
+            }
+          }
 
-          setInput(_input);
+          logger.trace('Processed input', _input);
         }
       } catch (error) {
         _error = error;
@@ -59,10 +69,6 @@ export function usePromptEnhancer() {
 
         setEnhancingPrompt(false);
         setPromptEnhanced(true);
-
-        setTimeout(() => {
-          setInput(_input);
-        });
       }
     }
   };

app/routes/api.enhancer.ts

@@ -1,54 +1,76 @@
 import { type ActionFunctionArgs } from '@remix-run/cloudflare';
-import { StreamingTextResponse, parseStreamPart } from 'ai';
 import { streamText } from '~/lib/.server/llm/stream-text';
 import { stripIndents } from '~/utils/stripIndent';
 
 const encoder = new TextEncoder();
-const decoder = new TextDecoder();
 
-export async function action(args: ActionFunctionArgs) {
-  return enhancerAction(args);
+// 模拟数据生成函数
+function* mockDataGenerator(message: string) {
+  const mockResponse = `您好！我希望开发一个功能完善的考勤管理系统。该系统应该包括以下功能：
+
+1. 员工签到和签退功能
+2. 请假和加班申请管理
+3. 考勤数据统计和报表生成
+4. 管理员后台管理界面
+5. 移动端适配，方便员工随时查看和操作
+6. 与公司现有人事系统的集成
+
+请问您能帮我设计这个系统的基本架构和主要功能模块吗？我希望使用React作为前端框架，Node.js作为后端，MongoDB作为数据库。同时，我也想了解一下如何确保系统的安全性和数据隐私保护。谢谢！`;
+
+  for (let i = 0; i < mockResponse.length; i += 2) {
+    yield mockResponse.slice(i, i + 2);
+  }
 }
 
-async function enhancerAction({ context, request }: ActionFunctionArgs) {
+// 环境变量或配置来控制是否使用模拟数据
+// const USE_MOCK_DATA = process.env.USE_MOCK_DATA === 'true';
+const USE_MOCK_DATA = true;
+
+export async function action({ context, request }: ActionFunctionArgs) {
   const { message } = await request.json<{ message: string }>();
 
   try {
-    const result = await streamText(
-      [
-        {
-          role: 'user',
-          content: stripIndents`
-          I want you to improve the user prompt that is wrapped in \`<original_prompt>\` tags.
+    if (USE_MOCK_DATA) {
+      // 使用模拟数据
+      const stream = new ReadableStream({
+        async start(controller) {
+          for (const chunk of mockDataGenerator(message)) {
+            controller.enqueue(encoder.encode(`0:${JSON.stringify(chunk)}\n`));
+            await new Promise(resolve => setTimeout(resolve, 50)); // 模拟延迟
+          }
+          controller.enqueue(encoder.encode('e:{"finishReason":"unknown","usage":{"promptTokens":null,"completionTokens":null},"isContinued":false}\n'));
+          controller.enqueue(encoder.encode('d:{"finishReason":"unknown","usage":{"promptTokens":null,"completionTokens":null}}\n'));
+          controller.close();
+        }
+      });
 
-          IMPORTANT: Only respond with the improved prompt and nothing else!
+      return new Response(stream, {
+        headers: { 'Content-Type': 'text/plain; charset=utf-8' }
+      });
+    } else {
+      // 使用实际 AI 服务
+      const result = await streamText(
+        [
+          {
+            role: 'user',
+            content: stripIndents`
+            I want you to improve the user prompt that is wrapped in \`<original_prompt>\` tags.
 
-          <original_prompt>
-            ${message}
-          </original_prompt>
-        `,
-        },
-      ],
-      context.cloudflare.env,
-    );
+            IMPORTANT: Only respond with the improved prompt and nothing else!
 
-    const transformStream = new TransformStream({
-      transform(chunk, controller) {
-        const processedChunk = decoder
-          .decode(chunk)
-          .split('\n')
-          .filter((line) => line !== '')
-          .map(parseStreamPart)
-          .map((part) => part.value)
-          .join('');
+            Also, please ensure your response is entirely in Chinese.
 
-        controller.enqueue(encoder.encode(processedChunk));
-      },
-    });
+            <original_prompt>
+              ${message}
+            </original_prompt>
+          `,
+          },
+        ],
+        context.cloudflare.env,
+      );
 
-    const transformedStream = result.toAIStream().pipeThrough(transformStream);
-
-    return new StreamingTextResponse(transformedStream);
+      return result.toDataStreamResponse();
+    }
   } catch (error) {
     console.log(error);