ChatGPT/FastGPT
1
0
Fork 0
mirror of https://github.com/labring/FastGPT.git synced 2025-07-23 05:12:39 +00:00
Code Issues Packages Projects Releases Wiki Activity

dataset save raw file

Browse Source
This commit is contained in:
archer
2023-09-03 22:39:09 +08:00
parent 086ea83fac
commit a754ceaf3b
37 changed files with 347 additions and 144 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
client/package.json
View File

@@ -31,6 +31,7 @@
"i18next": "^22.5.1",
"immer": "^9.0.19",
"js-cookie": "^3.0.5",
"jschardet": "^3.0.0",
"jsdom": "^22.1.0",
"jsonwebtoken": "^9.0.0",
"lodash": "^4.17.21",

10
client/pnpm-lock.yaml generated
View File

@@ -71,6 +71,9 @@ dependencies:
js-cookie:
specifier: ^3.0.5
version: registry.npmmirror.com/js-cookie@3.0.5
jschardet:
specifier: ^3.0.0
version: registry.npmmirror.com/jschardet@3.0.0
jsdom:
specifier: ^22.1.0
version: registry.npmmirror.com/jsdom@22.1.0
@@ -8918,6 +8921,13 @@ packages:
argparse: registry.npmmirror.com/argparse@2.0.1
dev: true
registry.npmmirror.com/jschardet@3.0.0:
resolution: {integrity: sha512-lJH6tJ77V8Nzd5QWRkFYCLc13a3vADkh3r/Fi8HupZGWk2OVVDfnZP8V/VgQgZ+lzW0kG2UGb5hFgt3V3ndotQ==, registry: https://registry.npm.taobao.org/, tarball: https://registry.npmmirror.com/jschardet/-/jschardet-3.0.0.tgz}
name: jschardet
version: 3.0.0
engines: {node: '>=0.1.90'}
dev: false
registry.npmmirror.com/jsdom@22.1.0:
resolution: {integrity: sha512-/9AVW7xNbsBv6GfWho4TTNjEo9fe6Zhf9O7s0Fhhr3u+awPwAJMKwAMXnkk5vBxflqLW9hTHX/0cs+P3gW+cQw==, registry: https://registry.npm.taobao.org/, tarball: https://registry.npmmirror.com/jsdom/-/jsdom-22.1.0.tgz}
name: jsdom

3
client/public/locales/en/common.json
View File

@@ -88,13 +88,16 @@
},
"file": {
"Click to download CSV template": "Click to download CSV template",
"Click to view file": "Click to view file",
"Create File": "Create File",
"Create file": "Create file",
"Drag and drop": "Drag and drop files here",
"Fetch Url": "Fetch Url",
"If the imported file is garbled, please convert CSV to UTF-8 encoding format": "If the imported file is garbled, please convert CSV to UTF-8 encoding format",
"Parse": "{{name}} Parsing...",
"Release the mouse to upload the file": "Release the mouse to upload the file",
"Select a maximum of 10 files": "Select a maximum of 10 files",
"Uploading": "Uploading: {{name}}, Progress: {{percent}}%",
"max 10": "Max 10 files",
"select a document": "select a document",
"support": "support {{fileExtension}} file",

3
client/public/locales/zh/common.json
View File

@@ -88,13 +88,16 @@
},
"file": {
"Click to download CSV template": "点击下载 CSV 模板",
"Click to view file": "点击查看原始文件",
"Create File": "创建新文件",
"Create file": "创建文件",
"Drag and drop": "拖拽文件至此",
"Fetch Url": "链接读取",
"If the imported file is garbled, please convert CSV to UTF-8 encoding format": "如果导入文件乱码，请将 CSV 转成 UTF-8 编码格式",
"Parse": "{{name}} 解析中...",
"Release the mouse to upload the file": "松开鼠标上传文件",
"Select a maximum of 10 files": "最多选择10个文件",
"Uploading": "正在上传 {{name}}，进度: {{percent}}%",
"max 10": "最多选择 10 个文件",
"select a document": "选择文件",
"support": "支持 {{fileExtension}} 文件",

11
client/src/api/plugins/kb.ts
View File

@@ -1,5 +1,5 @@
import { GET, POST, PUT, DELETE } from '../request';
import type { KbItemType, KbListItemType } from '@/types/plugin';
import type { DatasetItemType, KbItemType, KbListItemType } from '@/types/plugin';
import { RequestPaging } from '@/types/index';
import { TrainingModeEnum } from '@/constants/plugin';
import {
@@ -13,6 +13,7 @@ import {
import { Response as KbDataItemType } from '@/pages/api/plugins/kb/data/getDataById';
import { Props as UpdateDataProps } from '@/pages/api/openapi/kb/updateData';
import type { KbUpdateParams, CreateKbParams } from '../request/kb';
import { QuoteItemType } from '@/types/chat';
/* knowledge base */
export const getKbList = () => GET<KbListItemType[]>(`/plugins/kb/list`);
@@ -58,7 +59,7 @@ export const getTrainingData = (data: { kbId: string; init: boolean }) =>
export const getTrainingQueueLen = () => GET<number>(`/plugins/kb/data/getQueueLen`);
export const getKbDataItemById = (dataId: string) =>
GET<KbDataItemType>(`/plugins/kb/data/getDataById`, { dataId });
GET<QuoteItemType>(`/plugins/kb/data/getDataById`, { dataId });
/**
* 直接push数据
@@ -69,10 +70,8 @@ export const postKbDataFromList = (data: PushDataProps) =>
/**
* insert one data to dataset
*/
export const insertData2Kb = (data: {
kbId: string;
data: { a: string; q: string; source?: string };
}) => POST<string>(`/plugins/kb/data/insertData`, data);
export const insertData2Kb = (data: { kbId: string; data: DatasetItemType }) =>
POST<string>(`/plugins/kb/data/insertData`, data);
/**
* 更新一条数据

8
client/src/api/request.ts
View File

@@ -1,4 +1,9 @@
import axios, { Method, InternalAxiosRequestConfig, AxiosResponse } from 'axios';
import axios, {
Method,
InternalAxiosRequestConfig,
AxiosResponse,
AxiosProgressEvent
} from 'axios';
import { clearToken, getToken } from '@/utils/user';
import { TOKEN_ERROR_CODE } from '@/service/errorCode';
@@ -6,6 +11,7 @@ interface ConfigType {
headers?: { [key: string]: string };
hold?: boolean;
timeout?: number;
onUploadProgress?: (progressEvent: AxiosProgressEvent) => void;
}
interface ResponseDataType {
code: number;

14
client/src/api/system.ts
View File

@@ -1,6 +1,20 @@
import { GET, POST, PUT } from './request';
import type { InitDateResponse } from '@/pages/api/system/getInitData';
import { AxiosProgressEvent } from 'axios';
export const getInitData = () => GET<InitDateResponse>('/system/getInitData');
export const uploadImg = (base64Img: string) => POST<string>('/system/uploadImage', { base64Img });
export const postUploadFiles = (
data: FormData,
onUploadProgress: (progressEvent: AxiosProgressEvent) => void
) =>
POST<string[]>('/plugins/file/upload', data, {
onUploadProgress,
headers: {
'Content-Type': 'multipart/form-data; charset=utf-8'
}
});
export const getFileViewUrl = (fileId: string) => GET<string>('/plugins/file/readUrl', { fileId });

8
client/src/components/ChatBox/ContextModal.tsx
View File

@@ -21,7 +21,13 @@ const ContextModal = ({
minW={['90vw', '600px']}
isCentered
>
<ModalBody pt={0} whiteSpace={'pre-wrap'} textAlign={'justify'} fontSize={'sm'}>
<ModalBody
pt={0}
whiteSpace={'pre-wrap'}
textAlign={'justify'}
wordBreak={'break-all'}
fontSize={'sm'}
>
{context.map((item, i) => (
<Box
key={i}

46
client/src/components/ChatBox/QuoteModal.tsx
View File

@@ -6,15 +6,12 @@ import { useToast } from '@/hooks/useToast';
import { getErrText } from '@/utils/tools';
import { QuoteItemType } from '@/types/chat';
import MyIcon from '@/components/Icon';
import InputDataModal from '@/pages/kb/detail/components/InputDataModal';
import InputDataModal, { RawFileText } from '@/pages/kb/detail/components/InputDataModal';
import MyModal from '../MyModal';
import { KbDataItemType } from '@/types/plugin';
type SearchType = {
type SearchType = KbDataItemType & {
kb_id?: string;
id?: string;
q: string;
a?: string;
source?: string | undefined;
};
const QuoteModal = ({
@@ -29,12 +26,7 @@ const QuoteModal = ({
const theme = useTheme();
const { toast } = useToast();
const { setIsLoading, Loading } = useLoading();
const [editDataItem, setEditDataItem] = useState<{
kbId: string;
dataId: string;
a: string;
q: string;
}>();
const [editDataItem, setEditDataItem] = useState<QuoteItemType>();
/**
* click edit, get new kbDataItem
@@ -44,19 +36,14 @@ const QuoteModal = ({
if (!item.id) return;
try {
setIsLoading(true);
const data = (await getKbDataItemById(item.id)) as QuoteItemType;
const data = await getKbDataItemById(item.id);
if (!data) {
onUpdateQuote(item.id, '已删除');
throw new Error('该数据已被删除');
}
setEditDataItem({
kbId: data.kb_id,
dataId: data.id,
q: data.q,
a: data.a
});
setEditDataItem(data);
} catch (err) {
toast({
status: 'warning',
@@ -85,7 +72,13 @@ const QuoteModal = ({
</>
}
>
<ModalBody pt={0} whiteSpace={'pre-wrap'} textAlign={'justify'} fontSize={'sm'}>
<ModalBody
pt={0}
whiteSpace={'pre-wrap'}
textAlign={'justify'}
wordBreak={'break-all'}
fontSize={'sm'}
>
{rawSearch.map((item, i) => (
<Box
key={i}
@@ -98,7 +91,7 @@ const QuoteModal = ({
_hover={{ '& .edit': { display: 'flex' } }}
overflow={'hidden'}
>
{item.source && <Box color={'myGray.600'}>({item.source})</Box>}
{item.source && <RawFileText filename={item.source} fileId={item.file_id} />}
<Box>{item.q}</Box>
<Box>{item.a}</Box>
{item.id && (
@@ -136,10 +129,13 @@ const QuoteModal = ({
{editDataItem && (
<InputDataModal
onClose={() => setEditDataItem(undefined)}
onSuccess={() => onUpdateQuote(editDataItem.dataId, '手动修改')}
onDelete={() => onUpdateQuote(editDataItem.dataId, '已删除')}
kbId={editDataItem.kbId}
defaultValues={editDataItem}
onSuccess={() => onUpdateQuote(editDataItem.id, '手动修改')}
onDelete={() => onUpdateQuote(editDataItem.id, '已删除')}
kbId={editDataItem.kb_id}
defaultValues={{
...editDataItem,
dataId: editDataItem.id
}}
/>
)}
</>

10
client/src/components/Loading/index.tsx
View File

@@ -1,7 +1,7 @@
import React from 'react';
import { Spinner, Flex } from '@chakra-ui/react';
import { Spinner, Flex, Box } from '@chakra-ui/react';
const Loading = ({ fixed = true }: { fixed?: boolean }) => {
const Loading = ({ fixed = true, text = '' }: { fixed?: boolean; text?: string }) => {
return (
<Flex
position={fixed ? 'fixed' : 'absolute'}
@@ -13,8 +13,14 @@ const Loading = ({ fixed = true }: { fixed?: boolean }) => {
bottom={0}
alignItems={'center'}
justifyContent={'center'}
flexDirection={'column'}
>
<Spinner thickness="4px" speed="0.65s" emptyColor="myGray.100" color="myBlue.600" size="xl" />
{text && (
<Box mt={2} color="myBlue.700" fontWeight={'bold'}>
{text}
</Box>
)}
</Flex>
);
};

3
client/src/constants/common.ts
View File

@@ -15,7 +15,8 @@ export const fileImgs = [
export enum TrackEventName {
windowError = 'windowError',
pageError = 'pageError'
pageError = 'pageError',
wordReadError = 'wordReadError'
}
export const htmlTemplate = `<!DOCTYPE html>

12
client/src/hooks/useLoading.tsx
View File

@@ -5,8 +5,16 @@ export const useLoading = (props?: { defaultLoading: boolean }) => {
const [isLoading, setIsLoading] = useState(props?.defaultLoading || false);
const Loading = useCallback(
({ loading, fixed = true }: { loading?: boolean; fixed?: boolean }): JSX.Element | null => {
return isLoading || loading ? <LoadingComponent fixed={fixed} /> : null;
({
loading,
fixed = true,
text = ''
}: {
loading?: boolean;
fixed?: boolean;
text?: string;
}): JSX.Element | null => {
return isLoading || loading ? <LoadingComponent fixed={fixed} text={text} /> : null;
},
[isLoading]
);

35
client/src/pages/api/admin/initv43.ts Normal file
View File

@@ -0,0 +1,35 @@
// Next.js API route support: https://nextjs.org/docs/api-routes/introduction
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { authUser } from '@/service/utils/auth';
import { PgClient } from '@/service/pg';
import { PgTrainingTableName } from '@/constants/plugin';
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
try {
await authUser({ req, authRoot: true });
const { rowCount } = await PgClient.query(`SELECT 1
FROM information_schema.columns
WHERE table_schema = 'public'
AND table_name = '${PgTrainingTableName}'
AND column_name = 'file_id'`);
if (rowCount > 0) {
return jsonRes(res, {
data: '已经存在file_id字段'
});
}
jsonRes(res, {
data: await PgClient.query(
`ALTER TABLE ${PgTrainingTableName} ADD COLUMN file_id VARCHAR(100)`
)
});
} catch (error) {
jsonRes(res, {
code: 500,
error
});
}
}

28
client/src/pages/api/openapi/kb/pushData.ts
View File

@@ -9,12 +9,11 @@ import { startQueue } from '@/service/utils/tools';
import { PgClient } from '@/service/pg';
import { modelToolMap } from '@/utils/plugin';
import { getVectorModel } from '@/service/utils/data';
export type DateItemType = { a: string; q: string; source?: string };
import { DatasetItemType } from '@/types/plugin';
export type Props = {
kbId: string;
data: DateItemType[];
data: DatasetItemType[];
mode: `${TrainingModeEnum}`;
prompt?: string;
};
@@ -95,7 +94,7 @@ export async function pushDataToKb({
// 过滤重复的 qa 内容
const set = new Set();
const filterData: DateItemType[] = [];
const filterData: DatasetItemType[] = [];
data.forEach((item) => {
if (!item.q) return;
@@ -120,13 +119,10 @@ export async function pushDataToKb({
// 数据库去重
const insertData = (
await Promise.allSettled(
filterData.map(async ({ q, a = '', source }) => {
filterData.map(async (data) => {
let { q, a } = data;
if (mode !== TrainingModeEnum.index) {
return Promise.resolve({
q,
a,
source
});
return Promise.resolve(data);
}
if (!q) {
@@ -152,23 +148,17 @@ export async function pushDataToKb({
console.log(error);
error;
}
return Promise.resolve({
q,
a,
source
});
return Promise.resolve(data);
})
)
)
.filter((item) => item.status === 'fulfilled')
.map<DateItemType>((item: any) => item.value);
.map<DatasetItemType>((item: any) => item.value);
// 插入记录
const insertRes = await TrainingData.insertMany(
insertData.map((item) => ({
q: item.q,
a: item.a,
source: item.source,
...item,
userId,
kbId,
mode,

6
client/src/pages/api/openapi/kb/searchTest.ts
View File

@@ -41,7 +41,7 @@ export default withNextCors(async function handler(req: NextApiRequest, res: Nex
const response: any = await PgClient.query(
`BEGIN;
SET LOCAL ivfflat.probes = ${global.systemEnv.pgIvfflatProbe || 10};
select id,q,a,source,(vector <#> '[${
select id, q, a, source, file_id, (vector <#> '[${
vectors[0]
}]') * -1 AS score from ${PgTrainingTableName} where kb_id='${kbId}' AND user_id='${userId}' order by vector <#> '[${
vectors[0]
@@ -49,7 +49,9 @@ export default withNextCors(async function handler(req: NextApiRequest, res: Nex
COMMIT;`
);
jsonRes<Response>(res, { data: response?.[2]?.rows || [] });
jsonRes<Response>(res, {
data: response?.[2]?.rows || []
});
} catch (err) {
console.log(err);
jsonRes(res, {

10
client/src/pages/api/plugins/file/read.ts
View File

@@ -3,6 +3,7 @@ import { jsonRes } from '@/service/response';
import { connectToDatabase } from '@/service/mongo';
import { GridFSStorage } from '@/service/lib/gridfs';
import { authFileToken } from './readUrl';
import jschardet from 'jschardet';
export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
@@ -12,6 +13,10 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
const { fileId, userId } = await authFileToken(token);
if (!fileId) {
throw new Error('fileId is empty');
}
const gridFs = new GridFSStorage('dataset', userId);
const [file, buffer] = await Promise.all([
@@ -19,9 +24,12 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
gridFs.download(fileId)
]);
res.setHeader('encoding', file.encoding);
const encoding = jschardet.detect(buffer)?.encoding;
res.setHeader('encoding', encoding);
res.setHeader('Content-Type', file.contentType);
res.setHeader('Cache-Control', 'public, max-age=3600');
res.setHeader('Content-Disposition', `inline; filename="${encodeURIComponent(file.filename)}"`);
res.end(buffer);
} catch (error) {

16
client/src/pages/api/plugins/file/upload.ts
View File

@@ -28,9 +28,10 @@ class UploadModel {
limits: {
fieldSize: maxSize
},
preservePath: true,
storage: multer.diskStorage({
filename: (_req, file, cb) => {
const { ext } = path.parse(file.originalname);
const { ext } = path.parse(decodeURIComponent(file.originalname));
cb(null, nanoid() + ext);
}
})
@@ -44,8 +45,13 @@ class UploadModel {
return reject(error);
}
// @ts-ignore
resolve({ files: req.files });
resolve({
// @ts-ignore
files: req.files?.map((file) => ({
...file,
originalname: decodeURIComponent(file.originalname)
}))
});
});
});
}
@@ -56,9 +62,9 @@ const upload = new UploadModel();
export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
await connectToDatabase();
const { userId } = await authUser({ req });
const { userId } = await authUser({ req, authToken: true });
const { files } = await upload.doUpload(req, res);
const { files = [] } = await upload.doUpload(req, res);
const gridFs = new GridFSStorage('dataset', userId);

2
client/src/pages/api/plugins/kb/data/getDataById.ts
View File

@@ -30,7 +30,7 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
const where: any = [['user_id', userId], 'AND', ['id', dataId]];
const searchRes = await PgClient.select<KbDataItemType>(PgTrainingTableName, {
fields: ['kb_id', 'id', 'q', 'a', 'source'],
fields: ['kb_id', 'id', 'q', 'a', 'source', 'file_id'],
where,
limit: 1
});

2
client/src/pages/api/plugins/kb/data/getDataList.ts
View File

@@ -43,7 +43,7 @@ export default async function handler(req: NextApiRequest, res: NextApiResponse<
const [searchRes, total] = await Promise.all([
PgClient.select<KbDataItemType>(PgTrainingTableName, {
fields: ['id', 'q', 'a', 'source'],
fields: ['id', 'q', 'a', 'source', 'file_id'],
where,
order: [{ field: 'id', mode: 'DESC' }],
limit: pageSize,

3
client/src/pages/api/plugins/kb/data/insertData.ts
View File

@@ -8,10 +8,11 @@ import { insertKbItem, PgClient } from '@/service/pg';
import { modelToolMap } from '@/utils/plugin';
import { getVectorModel } from '@/service/utils/data';
import { getVector } from '@/pages/api/openapi/plugin/vector';
import { DatasetItemType } from '@/types/plugin';
export type Props = {
kbId: string;
data: { a: string; q: string; source?: string };
data: DatasetItemType;
};
export default withNextCors(async function handler(req: NextApiRequest, res: NextApiResponse<any>) {

3
client/src/pages/kb/detail/components/DataCard.tsx
View File

@@ -198,8 +198,7 @@ const DataCard = ({ kbId }: { kbId: string }) => {
onClick={() =>
setEditInputData({
dataId: item.id,
q: item.q,
a: item.a
...item
})
}
>

7
client/src/pages/kb/detail/components/Import/Chunk.tsx
View File

@@ -109,10 +109,9 @@ const ChunkImport = ({ kbId }: { kbId: string }) => {
return {
...file,
tokens: splitRes.tokens,
chunks: splitRes.chunks.map((chunk) => ({
q: chunk,
a: '',
source: file.filename
chunks: file.chunks.map((chunk, i) => ({
...chunk,
q: splitRes.chunks[i]
}))
};
})

3
client/src/pages/kb/detail/components/Import/Csv.tsx
View File

@@ -1,11 +1,10 @@
import React, { useState, useCallback, useMemo } from 'react';
import React, { useState, useMemo } from 'react';
import { Box, Flex, Button, useTheme, Image } from '@chakra-ui/react';
import { useToast } from '@/hooks/useToast';
import { useConfirm } from '@/hooks/useConfirm';
import { useMutation } from '@tanstack/react-query';
import { postKbDataFromList } from '@/api/plugins/kb';
import { getErrText } from '@/utils/tools';
import { vectorModelList } from '@/store/static';
import MyIcon from '@/components/Icon';
import DeleteIcon, { hoverDeleteStyles } from '@/components/Icon/delete';
import { TrainingModeEnum } from '@/constants/plugin';

70
client/src/pages/kb/detail/components/Import/FileSelect.tsx
View File

@@ -2,7 +2,13 @@ import MyIcon from '@/components/Icon';
import { useLoading } from '@/hooks/useLoading';
import { useSelectFile } from '@/hooks/useSelectFile';
import { useToast } from '@/hooks/useToast';
import { fileDownload, readCsvContent, simpleText, splitText2Chunks } from '@/utils/file';
import {
fileDownload,
readCsvContent,
simpleText,
splitText2Chunks,
uploadFiles
} from '@/utils/file';
import { Box, Flex, useDisclosure, type BoxProps } from '@chakra-ui/react';
import { fileImgs } from '@/constants/common';
import { DragEvent, useCallback, useState } from 'react';
@@ -11,7 +17,8 @@ import { readTxtContent, readPdfContent, readDocContent } from '@/utils/file';
import { customAlphabet } from 'nanoid';
import dynamic from 'next/dynamic';
import MyTooltip from '@/components/MyTooltip';
import { FetchResultItem } from '@/types/plugin';
import { FetchResultItem, DatasetItemType } from '@/types/plugin';
import { getErrText } from '@/utils/tools';
const UrlFetchModal = dynamic(() => import('./UrlFetchModal'));
const CreateFileModal = dynamic(() => import('./CreateFileModal'));
@@ -22,7 +29,7 @@ const csvTemplate = `question,answer,source\n"什么是 laf","laf 是一个云
export type FileItemType = {
id: string;
filename: string;
chunks: { q: string; a: string; source?: string }[];
chunks: DatasetItemType[];
text: string;
icon: string;
tokens: number;
@@ -58,7 +65,7 @@ const FileSelect = ({
});
const [isDragging, setIsDragging] = useState(false);
const [selecting, setSelecting] = useState(false);
const [selectingText, setSelectingText] = useState<string>();
const {
isOpen: isOpenUrlFetch,
@@ -73,7 +80,6 @@ const FileSelect = ({
const onSelectFile = useCallback(
async (files: File[]) => {
setSelecting(true);
try {
// Parse file by file
const chunkFiles: FileItemType[] = [];
@@ -88,19 +94,31 @@ const FileSelect = ({
continue;
}
let text = await (async () => {
switch (extension) {
case 'txt':
case 'md':
return readTxtContent(file);
case 'pdf':
return readPdfContent(file);
case 'doc':
case 'docx':
return readDocContent(file);
}
return '';
})();
// parse and upload files
let [text, filesId] = await Promise.all([
(async () => {
switch (extension) {
case 'txt':
case 'md':
return readTxtContent(file);
case 'pdf':
return readPdfContent(file);
case 'doc':
case 'docx':
return readDocContent(file);
}
return '';
})(),
uploadFiles(files, (percent) => {
if (percent < 100) {
setSelectingText(
t('file.Uploading', { name: file.name.slice(0, 20), percent }) || ''
);
} else {
setSelectingText(t('file.Parse', { name: file.name.slice(0, 20) }) || '');
}
})
]);
if (text) {
text = simpleText(text);
@@ -117,7 +135,8 @@ const FileSelect = ({
chunks: splitRes.chunks.map((chunk) => ({
q: chunk,
a: '',
source: file.name
source: file.name,
file_id: filesId[0]
}))
};
chunkFiles.unshift(fileItem);
@@ -139,7 +158,8 @@ const FileSelect = ({
chunks: data.map((item) => ({
q: item[0],
a: item[1],
source: item[2] || file.name
source: item[2] || file.name,
file_id: filesId[0]
}))
};
@@ -150,13 +170,13 @@ const FileSelect = ({
} catch (error: any) {
console.log(error);
toast({
title: typeof error === 'string' ? error : '解析文件失败',
title: getErrText(error, '解析文件失败'),
status: 'error'
});
}
setSelecting(false);
setSelectingText(undefined);
},
[chunkLen, onPushFiles, toast]
[chunkLen, onPushFiles, t, toast]
);
const onUrlFetch = useCallback(
(e: FetchResultItem[]) => {
@@ -353,7 +373,9 @@ const FileSelect = ({
{t('file.Click to download CSV template')}
</Box>
)}
<FileSelectLoading loading={selecting} fixed={false} />
{selectingText !== undefined && (
<FileSelectLoading loading text={selectingText} fixed={false} />
)}
<File onSelect={onSelectFile} />
{isOpenUrlFetch && <UrlFetchModal onClose={onCloseUrlFetch} onSuccess={onUrlFetch} />}
{isOpenCreateFile && <CreateFileModal onClose={onCloseCreateFile} onSuccess={onCreateFile} />}

7
client/src/pages/kb/detail/components/Import/QA.tsx
View File

@@ -97,10 +97,9 @@ const QAImport = ({ kbId }: { kbId: string }) => {
return {
...file,
tokens: splitRes.tokens,
chunks: splitRes.chunks.map((chunk) => ({
q: chunk,
a: '',
source: file.filename
chunks: file.chunks.map((chunk, i) => ({
...chunk,
q: splitRes.chunks[i]
}))
};
})

82
client/src/pages/kb/detail/components/InputDataModal.tsx
View File

@@ -1,7 +1,8 @@
import React, { useState, useCallback } from 'react';
import { Box, Flex, Button, Textarea, IconButton } from '@chakra-ui/react';
import { Box, Flex, Button, Textarea, IconButton, BoxProps } from '@chakra-ui/react';
import { useForm } from 'react-hook-form';
import { insertData2Kb, putKbDataById, delOneKbDataByDataId } from '@/api/plugins/kb';
import { getFileViewUrl } from '@/api/system';
import { useToast } from '@/hooks/useToast';
import { getErrText } from '@/utils/tools';
import MyIcon from '@/components/Icon';
@@ -10,8 +11,10 @@ import MyTooltip from '@/components/MyTooltip';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
import { useUserStore } from '@/store/user';
import { useQuery } from '@tanstack/react-query';
import { DatasetItemType } from '@/types/plugin';
import { useTranslation } from 'react-i18next';
export type FormData = { dataId?: string; a: string; q: string; source?: string };
export type FormData = { dataId?: string } & DatasetItemType;
const InputDataModal = ({
onClose,
@@ -29,12 +32,13 @@ const InputDataModal = ({
kbId: string;
defaultValues?: FormData;
}) => {
const { t } = useTranslation();
const [loading, setLoading] = useState(false);
const { toast } = useToast();
const { kbDetail, getKbDetail } = useUserStore();
const { register, handleSubmit, reset } = useForm<FormData>({
const { getValues, register, handleSubmit, reset } = useForm<FormData>({
defaultValues
});
@@ -183,7 +187,16 @@ const InputDataModal = ({
</Box>
</Box>
<Flex px={6} pt={2} pb={4} alignItems={'center'}>
<Flex px={6} pt={['34px', 2]} pb={4} alignItems={'center'} position={'relative'}>
<RawFileText
fileId={getValues('file_id')}
filename={getValues('source')}
position={'absolute'}
left={'50%'}
top={['16px', '50%']}
transform={'translate(-50%,-50%)'}
/>
<Box flex={1}>
{defaultValues.dataId && onDelete && (
<IconButton
@@ -217,15 +230,17 @@ const InputDataModal = ({
/>
)}
</Box>
<Button variant={'base'} mr={3} isLoading={loading} onClick={onClose}>
取消
</Button>
<Button
isLoading={loading}
onClick={handleSubmit(defaultValues.dataId ? updateData : sureImportData)}
>
{defaultValues.dataId ? '确认变更' : '确认导入'}
</Button>
<Box>
<Button variant={'base'} mr={3} isLoading={loading} onClick={onClose}>
取消
</Button>
<Button
isLoading={loading}
onClick={handleSubmit(defaultValues.dataId ? updateData : sureImportData)}
>
{defaultValues.dataId ? '确认变更' : '确认导入'}
</Button>
</Box>
</Flex>
</Flex>
</MyModal>
@@ -233,3 +248,44 @@ const InputDataModal = ({
};
export default InputDataModal;
interface RawFileTextProps extends BoxProps {
filename?: string;
fileId?: string;
}
export function RawFileText({ fileId, filename = '', ...props }: RawFileTextProps) {
const { t } = useTranslation();
const { toast } = useToast();
return (
<MyTooltip label={fileId ? t('file.Click to view file') || '' : ''} shouldWrapChildren={false}>
<Box
color={'myGray.600'}
display={'inline-block'}
{...(!!fileId
? {
cursor: 'pointer',
textDecoration: ['underline', 'none'],
_hover: {
textDecoration: 'underline'
},
onClick: async () => {
try {
const url = await getFileViewUrl(fileId);
const asPath = `${location.origin}${url}`;
window.open(asPath, '_blank');
} catch (error) {
toast({
title: getErrText(error, '获取文件地址失败'),
status: 'error'
});
}
}
}
: {})}
{...props}
>
{filename}
</Box>
</MyTooltip>
);
}

3
client/src/pages/kb/detail/components/Test.tsx
View File

@@ -207,8 +207,7 @@ const Test = ({ kbId }: { kbId: string }) => {
setEditData({
dataId: data.id,
q: data.q,
a: data.a
...data
});
} catch (err) {
toast({

5
client/src/service/events/generateQA.ts
View File

@@ -38,7 +38,7 @@ export async function generateQA(): Promise<any> {
prompt: 1,
q: 1,
source: 1,
model: 1
file_id: 1
});
// task preemption
@@ -136,7 +136,8 @@ A2:
kbId,
data: responseList.map((item) => ({
...item,
source: data.source
source: data.source,
file_id: data.file_id
})),
userId,
mode: TrainingModeEnum.index

2
client/src/service/events/generateVector.ts
View File

@@ -38,6 +38,7 @@ export async function generateVector(): Promise<any> {
q: 1,
a: 1,
source: 1,
file_id: 1,
vectorModel: 1
});
@@ -74,6 +75,7 @@ export async function generateVector(): Promise<any> {
q: dataItems[i].q,
a: dataItems[i].a,
source: data.source,
file_id: data.file_id,
vector
}))
});

4
client/src/service/models/trainingData.ts
View File

@@ -49,6 +49,10 @@ const TrainingDataSchema = new Schema({
source: {
type: String,
default: ''
},
file_id: {
type: String,
default: ''
}
});

2
client/src/service/moduleDispatch/kb/search.ts
View File

@@ -42,7 +42,7 @@ export async function dispatchKBSearch(props: Record<string, any>): Promise<KBSe
const res: any = await PgClient.query(
`BEGIN;
SET LOCAL ivfflat.probes = ${global.systemEnv.pgIvfflatProbe || 10};
select kb_id,id,q,a,source from ${PgTrainingTableName} where kb_id IN (${kbList
select kb_id,id,q,a,source,file_id from ${PgTrainingTableName} where kb_id IN (${kbList
.map((item) => `'${item.kbId}'`)
.join(',')}) AND vector <#> '[${vectors[0]}]' < -${similarity} order by vector <#> '[${
vectors[0]

17
client/src/service/pg.ts
View File

@@ -1,8 +1,8 @@
import { Pool } from 'pg';
import type { QueryResultRow } from 'pg';
import { PgTrainingTableName } from '@/constants/plugin';
import { exit } from 'process';
import { addLog } from './utils/tools';
import { DatasetItemType } from '@/types/plugin';
export const connectPg = async (): Promise<Pool> => {
if (global.pgClient) {
@@ -45,7 +45,7 @@ type DeleteProps = {
where: WhereProps;
};
type ValuesProps = { key: string; value: string | number }[];
type ValuesProps = { key: string; value?: string | number }[];
type UpdateProps = {
values: ValuesProps;
where: WhereProps;
@@ -168,18 +168,16 @@ export const insertKbItem = ({
}: {
userId: string;
kbId: string;
data: {
data: (DatasetItemType & {
vector: number[];
q: string;
a: string;
source?: string;
}[];
})[];
}) => {
return PgClient.insert(PgTrainingTableName, {
values: data.map((item) => [
{ key: 'user_id', value: userId },
{ key: 'kb_id', value: kbId },
{ key: 'source', value: item.source?.slice(0, 30)?.trim() || '' },
{ key: 'file_id', value: item.file_id },
{ key: 'q', value: item.q.replace(/'/g, '"') },
{ key: 'a', value: item.a.replace(/'/g, '"') },
{ key: 'vector', value: `[${item.vector}]` }
@@ -196,10 +194,11 @@ export async function initPg() {
id BIGSERIAL PRIMARY KEY,
vector VECTOR(1536) NOT NULL,
user_id VARCHAR(50) NOT NULL,
kb_id VARCHAR(50) NOT NULL,
kb_id VARCHAR(50),
source VARCHAR(100),
file_id VARCHAR(100),
q TEXT NOT NULL,
a TEXT NOT NULL
a TEXT
);
CREATE INDEX IF NOT EXISTS modelData_userId_index ON ${PgTrainingTableName} USING HASH (user_id);
CREATE INDEX IF NOT EXISTS modelData_kbId_index ON ${PgTrainingTableName} USING HASH (kb_id);

7
client/src/types/chat.d.ts vendored
View File

@@ -3,6 +3,7 @@ import type { InitChatResponse, InitShareChatResponse } from '@/api/response/cha
import { TaskResponseKeyEnum } from '@/constants/chat';
import { ClassifyQuestionAgentItemType } from './app';
import { ChatItemSchema } from './mongoSchema';
import { KbDataItemType } from './plugin';
export type ExportChatType = 'md' | 'pdf' | 'html';
@@ -41,12 +42,8 @@ export type ShareChatType = InitShareChatResponse & {
history: ShareChatHistoryItemType;
};
export type QuoteItemType = {
export type QuoteItemType = KbDataItemType & {
kb_id: string;
id: string;
q: string;
a: string;
source?: string;
};
export type ChatHistoryItemResType = {

1
client/src/types/mongoSchema.d.ts vendored
View File

@@ -78,6 +78,7 @@ export interface TrainingDataSchema {
q: string;
a: string;
source: string;
file_id: string;
}
export interface ChatSchema {

0
client/src/types/pg.d.ts vendored
View File

11
client/src/types/plugin.d.ts vendored
View File

@@ -20,12 +20,15 @@ export interface KbItemType {
tags: string;
}
export interface KbDataItemType {
id: string;
export type DatasetItemType = {
q: string; // 提问词
a: string; // 原文
source: string;
}
source?: string;
file_id?: string;
};
export type KbDataItemType = DatasetItemType & {
id: string;
};
export type KbTestItemType = {
id: string;

36
client/src/utils/file.ts
View File

@@ -2,7 +2,23 @@ import mammoth from 'mammoth';
import Papa from 'papaparse';
import { getOpenAiEncMap } from './plugin/openai';
import { getErrText } from './tools';
import { uploadImg } from '@/api/system';
import { uploadImg, postUploadFiles } from '@/api/system';
/**
* upload file to mongo gridfs
*/
export const uploadFiles = (files: File[], percentListen?: (percent: number) => void) => {
const form = new FormData();
files.forEach((file) => {
form.append('file', file, encodeURIComponent(file.name));
});
return postUploadFiles(form, (e) => {
if (!e.total) return;
const percent = Math.round((e.loaded / e.total) * 100);
percentListen && percentListen(percent);
});
};
/**
* 读取 txt 文件内容
@@ -37,7 +53,11 @@ export const readPdfContent = (file: File) =>
const readPDFPage = async (doc: any, pageNo: number) => {
const page = await doc.getPage(pageNo);
const tokenizedText = await page.getTextContent();
const pageText = tokenizedText.items.map((token: any) => token.str).join(' ');
const pageText = tokenizedText.items
.map((token: any) => token.str)
.filter((item: string) => item)
.join('');
return pageText;
};
@@ -54,12 +74,12 @@ export const readPdfContent = (file: File) =>
const pageTexts = await Promise.all(pageTextPromises);
resolve(pageTexts.join('\n'));
} catch (err) {
console.log(err, 'pdfjs error');
console.log(err, 'pdf load error');
reject('解析 PDF 失败');
}
};
reader.onerror = (err) => {
console.log(err, 'reader error');
console.log(err, 'pdf load error');
reject('解析 PDF 失败');
};
} catch (error) {
@@ -83,10 +103,18 @@ export const readDocContent = (file: File) =>
});
resolve(res?.value);
} catch (error) {
window.umami?.track('wordReadError', {
err: error?.toString()
});
console.log('error doc read:', error);
reject('读取 doc 文件失败, 请转换成 PDF');
}
};
reader.onerror = (err) => {
window.umami?.track('wordReadError', {
err: err?.toString()
});
console.log('error doc read:', err);
reject('读取 doc 文件失败');