perf: generate queue

src/pages/api/openapi/kb/pushData.ts

@@ -3,19 +3,21 @@ import type { KbDataItemType } from '@/types/plugin';
 import { jsonRes } from '@/service/response';
 import { connectToDatabase, TrainingData } from '@/service/mongo';
 import { authUser } from '@/service/utils/auth';
-import { generateVector } from '@/service/events/generateVector';
-import { PgClient } from '@/service/pg';
 import { authKb } from '@/service/utils/auth';
 import { withNextCors } from '@/service/utils/tools';
+import { TrainingTypeEnum } from '@/constants/plugin';
+import { startQueue } from '@/service/utils/tools';
 
-interface Props {
+export type Props = {
   kbId: string;
   data: { a: KbDataItemType['a']; q: KbDataItemType['q'] }[];
-}
+  mode: `${TrainingTypeEnum}`;
+  prompt?: string;
+};
 
 export default withNextCors(async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
   try {
-    const { kbId, data } = req.body as Props;
+    const { kbId, data, mode, prompt } = req.body as Props;
 
     if (!kbId || !Array.isArray(data)) {
       throw new Error('缺少参数');
@@ -29,7 +31,9 @@ export default withNextCors(async function handler(req: NextApiRequest, res: Nex
       data: await pushDataToKb({
         kbId,
         data,
-        userId
+        userId,
+        mode,
+        prompt
       })
     });
   } catch (err) {
@@ -40,36 +44,43 @@ export default withNextCors(async function handler(req: NextApiRequest, res: Nex
   }
 });
 
-export async function pushDataToKb({ userId, kbId, data }: { userId: string } & Props) {
+export async function pushDataToKb({
+  userId,
+  kbId,
+  data,
+  mode,
+  prompt
+}: { userId: string } & Props) {
   await authKb({
     userId,
     kbId
   });
 
   if (data.length === 0) {
-    return {
-      trainingId: ''
-    };
+    return {};
   }
 
   // 插入记录
-  const { _id } = await TrainingData.create({
-    userId,
-    kbId,
-    vectorList: data
-  });
+  await TrainingData.insertMany(
+    data.map((item) => ({
+      q: item.q,
+      a: item.a,
+      userId,
+      kbId,
+      mode,
+      prompt
+    }))
+  );
 
-  generateVector(_id);
+  startQueue();
 
-  return {
-    trainingId: _id
-  };
+  return {};
 }
 
 export const config = {
   api: {
     bodyParser: {
-      sizeLimit: '100mb'
+      sizeLimit: '20mb'
     }
   }
 };

src/pages/api/openapi/kb/updateData.ts

@@ -33,7 +33,15 @@ export default withNextCors(async function handler(req: NextApiRequest, res: Nex
     // 更新 pg 内容.仅修改a，不需要更新向量。
     await PgClient.update('modelData', {
       where: [['id', dataId], 'AND', ['user_id', userId]],
-      values: [{ key: 'a', value: a }, ...(q ? [{ key: 'q', value: `${vector[0]}` }] : [])]
+      values: [
+        { key: 'a', value: a },
+        ...(q
+          ? [
+              { key: 'q', value: q },
+              { key: 'vector', value: `[${vector[0]}]` }
+            ]
+          : [])
+      ]
     });
 
     jsonRes(res);

src/pages/api/openapi/text/splitText.ts

@@ -1,69 +0,0 @@
-import type { NextApiRequest, NextApiResponse } from 'next';
-import { jsonRes } from '@/service/response';
-import { connectToDatabase, TrainingData } from '@/service/mongo';
-import { authKb, authUser } from '@/service/utils/auth';
-import { generateQA } from '@/service/events/generateQA';
-import { TrainingTypeEnum } from '@/constants/plugin';
-import { withNextCors } from '@/service/utils/tools';
-import { pushDataToKb } from '../kb/pushData';
-
-/* split text */
-export default withNextCors(async function handler(req: NextApiRequest, res: NextApiResponse) {
-  try {
-    const { chunks, kbId, prompt, mode } = req.body as {
-      kbId: string;
-      chunks: string[];
-      prompt: string;
-      mode: `${TrainingTypeEnum}`;
-    };
-    if (!chunks || !kbId || !prompt) {
-      throw new Error('参数错误');
-    }
-    await connectToDatabase();
-
-    const { userId } = await authUser({ req });
-
-    // 验证是否是该用户的 model
-    await authKb({
-      kbId,
-      userId
-    });
-
-    if (mode === TrainingTypeEnum.qa) {
-      // 批量QA拆分插入数据
-      const { _id } = await TrainingData.create({
-        userId,
-        kbId,
-        qaList: chunks,
-        prompt
-      });
-      generateQA(_id);
-    } else if (mode === TrainingTypeEnum.subsection) {
-      // 分段导入，直接插入向量队列
-      const response = await pushDataToKb({
-        kbId,
-        data: chunks.map((item) => ({ q: item, a: '' })),
-        userId
-      });
-
-      return jsonRes(res, {
-        data: response
-      });
-    }
-
-    jsonRes(res);
-  } catch (err) {
-    jsonRes(res, {
-      code: 500,
-      error: err
-    });
-  }
-});
-
-export const config = {
-  api: {
-    bodyParser: {
-      sizeLimit: '100mb'
-    }
-  }
-};