perf: redirect request and err log replace (#768)

perf: dataset openapi

openapi

docSite/content/docs/development/openapi/dataset.md

@@ -342,7 +342,7 @@ data 为集合的 ID。
 {{< /tabs >}}
 
 
-### 创建一个纯文本集合(商业版)
+### 创建一个纯文本集合
 
 传入一段文字，创建一个集合，会根据传入的文字进行分割。
 
@@ -351,7 +351,7 @@ data 为集合的 ID。
 {{< markdownify >}}
 
 ```bash
-curl --location --request POST 'http://localhost:3000/api/proApi/core/dataset/collection/create/text' \
+curl --location --request POST 'http://localhost:3000/api/core/dataset/collection/create/text' \
 --header 'Authorization: Bearer {{authorization}}' \
 --header 'Content-Type: application/json' \
 --data-raw '{
@@ -418,7 +418,7 @@ data 为集合的 ID。
 {{< /tab >}}
 {{< /tabs >}}
 
-### 创建一个链接集合(商业版)
+### 创建一个链接集合
 
 传入一个网络链接，创建一个集合，会先去对应网页抓取内容，再抓取的文字进行分割。
 
@@ -427,7 +427,7 @@ data 为集合的 ID。
 {{< markdownify >}}
 
 ```bash
-curl --location --request POST 'http://localhost:3000/api/proApi/core/dataset/collection/create/link' \
+curl --location --request POST 'http://localhost:3000/api/core/dataset/collection/create/link' \
 --header 'Authorization: Bearer {{authorization}}' \
 --header 'Content-Type: application/json' \
 --data-raw '{

projects/app/src/pages/api/core/dataset/collection/create/link.ts

@@ -0,0 +1,88 @@
+/* 
+    Create one dataset collection
+*/
+import type { NextApiRequest, NextApiResponse } from 'next';
+import { jsonRes } from '@fastgpt/service/common/response';
+import { connectToDatabase } from '@/service/mongo';
+import type { LinkCreateDatasetCollectionParams } from '@fastgpt/global/core/dataset/api.d';
+import { authDataset } from '@fastgpt/service/support/permission/auth/dataset';
+import { createOneCollection } from '@fastgpt/service/core/dataset/collection/controller';
+import {
+  TrainingModeEnum,
+  DatasetCollectionTypeEnum
+} from '@fastgpt/global/core/dataset/constants';
+import { checkDatasetLimit } from '@fastgpt/service/support/permission/limit/dataset';
+import { predictDataLimitLength } from '@fastgpt/global/core/dataset/utils';
+import { createTrainingBill } from '@fastgpt/service/support/wallet/bill/controller';
+import { BillSourceEnum } from '@fastgpt/global/support/wallet/bill/constants';
+import { getQAModel, getVectorModel } from '@/service/core/ai/model';
+import { reloadCollectionChunks } from '@fastgpt/service/core/dataset/collection/utils';
+
+export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
+  try {
+    await connectToDatabase();
+    const {
+      link,
+      trainingType = TrainingModeEnum.chunk,
+      chunkSize = 512,
+      chunkSplitter,
+      qaPrompt,
+      ...body
+    } = req.body as LinkCreateDatasetCollectionParams;
+
+    const { teamId, tmbId, dataset } = await authDataset({
+      req,
+      authToken: true,
+      authApiKey: true,
+      datasetId: body.datasetId,
+      per: 'w'
+    });
+
+    // 1. check dataset limit
+    await checkDatasetLimit({
+      teamId,
+      freeSize: global.feConfigs?.subscription?.datasetStoreFreeSize,
+      insertLen: predictDataLimitLength(trainingType, new Array(10))
+    });
+
+    // 2. create collection
+    const collectionId = await createOneCollection({
+      ...body,
+      name: link,
+      teamId,
+      tmbId,
+      type: DatasetCollectionTypeEnum.link,
+
+      trainingType,
+      chunkSize,
+      chunkSplitter,
+      qaPrompt,
+
+      rawLink: link
+    });
+
+    // 3. create bill and start sync
+    const { billId } = await createTrainingBill({
+      teamId,
+      tmbId,
+      appName: 'core.dataset.collection.Sync Collection',
+      billSource: BillSourceEnum.training,
+      vectorModel: getVectorModel(dataset.vectorModel).name,
+      agentModel: getQAModel(dataset.agentModel).name
+    });
+    await reloadCollectionChunks({
+      collectionId,
+      tmbId,
+      billId
+    });
+
+    jsonRes(res, {
+      data: { collectionId }
+    });
+  } catch (err) {
+    jsonRes(res, {
+      code: 500,
+      error: err
+    });
+  }
+}

projects/app/src/pages/api/core/dataset/collection/create/text.ts

@@ -0,0 +1,117 @@
+/* 
+    Create one dataset collection
+*/
+import type { NextApiRequest, NextApiResponse } from 'next';
+import { jsonRes } from '@fastgpt/service/common/response';
+import { connectToDatabase } from '@/service/mongo';
+import type { TextCreateDatasetCollectionParams } from '@fastgpt/global/core/dataset/api.d';
+import { authDataset } from '@fastgpt/service/support/permission/auth/dataset';
+import { createOneCollection } from '@fastgpt/service/core/dataset/collection/controller';
+import {
+  TrainingModeEnum,
+  DatasetCollectionTypeEnum
+} from '@fastgpt/global/core/dataset/constants';
+import { splitText2Chunks } from '@fastgpt/global/common/string/textSplitter';
+import { checkDatasetLimit } from '@fastgpt/service/support/permission/limit/dataset';
+import { predictDataLimitLength } from '@fastgpt/global/core/dataset/utils';
+import { pushDataToTrainingQueue } from '@/service/core/dataset/data/controller';
+import { hashStr } from '@fastgpt/global/common/string/tools';
+import { createTrainingBill } from '@fastgpt/service/support/wallet/bill/controller';
+import { BillSourceEnum } from '@fastgpt/global/support/wallet/bill/constants';
+import { getQAModel, getVectorModel } from '@/service/core/ai/model';
+
+export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
+  try {
+    await connectToDatabase();
+    const {
+      name,
+      text,
+      trainingType = TrainingModeEnum.chunk,
+      chunkSize = 512,
+      chunkSplitter,
+      qaPrompt,
+      ...body
+    } = req.body as TextCreateDatasetCollectionParams;
+
+    const { teamId, tmbId, dataset } = await authDataset({
+      req,
+      authToken: true,
+      authApiKey: true,
+      datasetId: body.datasetId,
+      per: 'w'
+    });
+
+    // 1. split text to chunks
+    const { chunks } = splitText2Chunks({
+      text,
+      chunkLen: chunkSize,
+      overlapRatio: trainingType === TrainingModeEnum.chunk ? 0.2 : 0,
+      customReg: chunkSplitter ? [chunkSplitter] : []
+    });
+
+    // 2. check dataset limit
+    await checkDatasetLimit({
+      teamId,
+      freeSize: global.feConfigs?.subscription?.datasetStoreFreeSize,
+      insertLen: predictDataLimitLength(trainingType, chunks)
+    });
+
+    // 3. create collection and training bill
+    const [collectionId, { billId }] = await Promise.all([
+      createOneCollection({
+        ...body,
+        teamId,
+        tmbId,
+        type: DatasetCollectionTypeEnum.virtual,
+
+        name,
+        trainingType,
+        chunkSize,
+        chunkSplitter,
+        qaPrompt,
+
+        hashRawText: hashStr(text),
+        rawTextLength: text.length
+      }),
+      createTrainingBill({
+        teamId,
+        tmbId,
+        appName: name,
+        billSource: BillSourceEnum.training,
+        vectorModel: getVectorModel(dataset.vectorModel)?.name,
+        agentModel: getQAModel(dataset.agentModel)?.name
+      })
+    ]);
+
+    // 4. push chunks to training queue
+    const insertResults = await pushDataToTrainingQueue({
+      teamId,
+      tmbId,
+      collectionId,
+      trainingMode: trainingType,
+      prompt: qaPrompt,
+      billId,
+      data: chunks.map((text, index) => ({
+        q: text,
+        chunkIndex: index
+      }))
+    });
+
+    jsonRes(res, {
+      data: { collectionId, results: insertResults }
+    });
+  } catch (err) {
+    jsonRes(res, {
+      code: 500,
+      error: err
+    });
+  }
+}
+
+export const config = {
+  api: {
+    bodyParser: {
+      sizeLimit: '10mb'
+    }
+  }
+};

projects/app/src/web/core/dataset/api.ts

@@ -76,7 +76,7 @@ export const getDatasetCollectionById = (id: string) =>
 export const postDatasetCollection = (data: CreateDatasetCollectionParams) =>
   POST<string>(`/core/dataset/collection/create`, data);
 export const postCreateDatasetLinkCollection = (data: LinkCreateDatasetCollectionParams) =>
-  POST<{ collectionId: string }>(`/proApi/core/dataset/collection/create/link`, data);
+  POST<{ collectionId: string }>(`/core/dataset/collection/create/link`, data);
 
 export const putDatasetCollectionById = (data: UpdateDatasetCollectionParams) =>
   POST(`/core/dataset/collection/update`, data);