Compare commits

..

154 Commits

Author SHA1 Message Date
Archer
661ee79943 fix: CQ module output (#445) 2023-10-30 16:45:36 +08:00
Archer
60ee160131 v4.5.2 (#439) 2023-10-30 13:26:42 +08:00
Archer
008d0af010 Quote Modal UI and fix doc (#432) 2023-10-25 20:13:32 +08:00
lizhuang
f2fb0aedfd Update README.md 文档改为https://doc.fastgpt.in网站访问 (#424)
文档改为https://doc.fastgpt.in网站访问
2023-10-24 18:07:30 +08:00
Archer
1dca5edcc6 v4.5.1-3 (#427) 2023-10-24 17:32:36 +08:00
Archer
1942cb0d67 perf: btn color (#423) 2023-10-24 13:19:23 +08:00
Archer
bf6dbfb245 v4.5.1-2 (#421) 2023-10-23 15:05:13 +08:00
Archer
d37433eacd Config file to set doc baseurl (#419) 2023-10-23 08:56:43 +08:00
Archer
a3534407bf v4.5.1 (#417) 2023-10-22 23:54:04 +08:00
Carson Yang
3091a90df6 Update README (#418)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-22 23:47:09 +08:00
Carson Yang
41b8f4443c Docs: update qr for wechat group (#416)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-22 23:28:46 +08:00
Carson Yang
777f089423 Docs: update README (#407)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-18 15:51:51 +08:00
不做了睡大觉
b23e00f3e5 添加Baichuan2-7B-Chat模型接口文件 (#404)
* 更新镜像

* 更新镜像信息

* 更新镜像信息

* Create openai_api.py

* Create requirements.txt
2023-10-18 10:34:22 +08:00
Archer
3b776b6639 v4.5 (#403) 2023-10-17 10:00:32 +08:00
Archer
dd8f2744bf Extraction schema (#398) 2023-10-14 23:02:01 +08:00
左风
7db8d3ea0f Docs: add quick start and video link (#395) 2023-10-13 20:16:18 +08:00
Archer
ad7a17bf40 Optimize the project structure and introduce DDD design (#394) 2023-10-12 17:46:37 +08:00
李启爱
76ac5238b6 Update 447.md (#392) 2023-10-12 14:56:18 +08:00
Carson Yang
add73aa2c5 Docs: use docsearch (#391)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-12 00:04:45 +08:00
Archer
bcf9491999 v4.4.7-2 (#388) 2023-10-11 17:18:43 +08:00
Archer
d0041a98b4 Optimize the file storage structure of the knowledge base (#386) 2023-10-10 22:41:05 +08:00
Carson Yang
29d152784f Docs: delete image cdn for vercel (#385)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-09 15:03:07 +08:00
Carson Yang
cd7214ba8d Docs: update workflow for building docs image (#384)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-09 14:32:47 +08:00
Archer
6a84e73a82 fix: packages (#378) 2023-10-08 09:59:05 +08:00
Archer
98ce5103a0 v4.4.6 (#377) 2023-10-07 18:02:20 +08:00
Carson Yang
c65a36d3ab Docs: hide button for questionnaire on mobile device (#376)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-07 14:57:26 +08:00
Carson Yang
b6e49da288 Docs: update button for questionnaire (#375)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-06 23:52:45 +08:00
Archer
45998f9cf5 README (#372) 2023-10-06 21:19:44 +08:00
Carson Yang
4197f63751 Update README (#371)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-06 14:07:37 +08:00
Carson Yang
ace8134a16 Docs: add Dockerfile for docs (#369)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-06 08:01:16 +08:00
Carson Yang
7f1fecb84e Docs: update theme (#368)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-10-04 22:25:07 +08:00
Archer
bf172fab81 perf: markdown more wrap (#365) 2023-10-02 20:19:09 +08:00
Archer
36f5648cae perf: v4.4.6-1 (#364) 2023-09-28 17:30:05 +08:00
Archer
ab57bfcc4a perf: completions api.fix: new chat question guide (#361) 2023-09-27 12:05:13 +08:00
Archer
11848b8f44 v4.4.5-3 (#357) 2023-09-26 21:17:13 +08:00
epoh
a11e0bd9c3 Update chatglm2.md (#354) 2023-09-26 15:06:38 +08:00
Archer
f6552d0d4f v4.4.5-2 (#355) 2023-09-26 14:31:37 +08:00
epoh
38d4db5d5f Rename requirement.txt to requirements.txt (#352) 2023-09-26 09:38:14 +08:00
Archer
63cd379682 Add share link hook (#351) 2023-09-25 23:12:42 +08:00
Archer
9136c9306a Add OpenAPI docs;Correct the glm document (#346) 2023-09-25 14:28:44 +08:00
Byte Sound
c9db9f33ea Update intro.md (#348)
错别字,市区改为时区
2023-09-25 13:33:30 +08:00
Archer
3d7178d06f monorepo packages (#344) 2023-09-24 18:02:09 +08:00
Archer
a4ff5a3f73 perf: api key (#342) 2023-09-23 20:28:03 +08:00
Archer
814c5b3d3c Add bill of training and rate of file upload (#339) 2023-09-21 21:02:44 +08:00
Chen X
e7e0677291 Docs:add-workflow-case-全能助手 (#334) 2023-09-21 15:57:42 +08:00
Archer
823f4b7ad1 Optimize the structure and naming of projects (#335) 2023-09-21 14:49:56 +08:00
Carson Yang
a3c77480f7 Add action for translating Non-English issues content to English (#333)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-21 14:19:54 +08:00
Archer
e367265dbb feat: function call prompt version (#331) 2023-09-21 12:27:48 +08:00
Archer
7e0deb29e0 Add SSE controller; fix share page login failed (#330) 2023-09-20 16:34:32 +08:00
Archer
0d94db4331 fix: ts and default dataset (#329) 2023-09-20 11:43:49 +08:00
Carson Yang
177482b33a Docs: fix code block highlight (#328)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-20 11:43:35 +08:00
Archer
63b183a9fe fix: mark modal cannot select folder (#327) 2023-09-20 11:26:17 +08:00
Carson Yang
858117f8c0 Docs: update font to LXGW WenKai (#325)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-19 21:22:04 +08:00
Archer
ac4355d2e1 Add modal to show completion response data (#324) 2023-09-19 20:31:45 +08:00
Archer
ce7da2db66 Optimize chat reponse data (#322) 2023-09-19 16:10:30 +08:00
Archer
0a4a1def1e fix: connected error (#318) 2023-09-19 07:54:50 +08:00
Archer
35f4deca76 Revert "Feature: 高级编排自动布局 (#314)" (#319)
This reverts commit ba1451a0e9.
2023-09-18 23:44:44 +08:00
jaden
ba1451a0e9 Feature: 高级编排自动布局 (#314)
* feat: adFlow auto layout

* chore: delete file and build pnpm lock file
2023-09-18 23:39:19 +08:00
Archer
40d69e6e20 version (#317) 2023-09-18 21:56:38 +08:00
Sr
b8ba947ba8 feat: Added defaultOpen Attribute for iframe (#302)
* feat: Added defaultOpen Attribute for iframe

This commit introduces a new attribute `defaultOpen` for the iframe created in `iframe.js`. The `defaultOpen` attribute allows the iframe to be visible by default when the page loads. This new feature enhances the user experience by providing an option to display the chatbot window immediately after the page is loaded, without requiring user interaction.

* Update iframe.js

code standard
2023-09-18 21:27:08 +08:00
Archer
06be57815e v4.4.3 (#316) 2023-09-18 21:26:42 +08:00
Carson Yang
81e37a5736 Update architecture diagram (#315)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-18 21:26:15 +08:00
Archer
b8ea546b3f v4.2.2 (#312) 2023-09-18 13:37:25 +08:00
Carson Yang
0bb31b985d Docs: update style (#310)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-17 15:06:25 +08:00
Carson Yang
453824260f Docs: fix typo (#307)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-15 22:44:04 +08:00
hehan
a8fdffc3e9 Docs: intergate feishu (#305) 2023-09-15 14:32:43 +08:00
Carson Yang
24164d9454 Update deploy-docs-preview workflow (#304)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-15 13:43:36 +08:00
Archer
4365a94ea9 System optimize (#303) 2023-09-15 10:21:46 +08:00
Carson Yang
7c1ec04380 Docs: add github badge (#301)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-14 17:36:51 +08:00
Archer
09b6365321 perf: action cache (#300) 2023-09-13 22:17:55 +08:00
Archer
eb2e383cc7 perf: document icon and language select (#299) 2023-09-13 19:54:29 +08:00
Archer
ae4c479f37 file name (#297) 2023-09-13 18:23:55 +08:00
Archer
6a996272da fix: share link quote (#296) 2023-09-13 18:15:22 +08:00
Carson Yang
1bf76ebe7a Docs: add limiting responsibility (#295)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-13 17:00:46 +08:00
Archer
a19afca148 v4.4.1 (#294)
* move file

* perf: dataset file manage

* v441 description

* fix: qa csv update file

* feat: rename file

* frontend show system-version
2023-09-13 17:00:17 +08:00
Carson Yang
be3b680bc6 Docs: add community (#293)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-13 13:55:50 +08:00
Carson Yang
31dbcfde9f Docs: update cdn (#291)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-13 09:29:17 +08:00
Archer
6d438aafdf google login and power share link (#292) 2023-09-13 08:49:22 +08:00
Carson Yang
1aaafcf631 Docs: update weight (#290)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-12 22:36:34 +08:00
Carson Yang
7521bce77e Docs: update cdn (#289)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-12 21:33:12 +08:00
Carson Yang
c8dee29dc4 Docs: add pricing doc (#287)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-12 20:06:09 +08:00
Carson Yang
8f953d1fc4 Update README.md (#283)
Add demo video

Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-12 12:46:29 +08:00
Carson Yang
970b62be25 Docs: enable ‘Edit this page’ (#280)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-11 23:59:14 +08:00
Carson Yang
b2b3aa651d Docs: add details shortcode (#279)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-11 20:48:14 +08:00
Archer
b0e7d25464 docs weight (#278) 2023-09-11 18:36:43 +08:00
Archer
b46048609c feat: move dataset (#277) 2023-09-11 18:23:51 +08:00
Archer
ae2887e956 fix: file_id undefined bug (#275) 2023-09-11 10:15:52 +08:00
Archer
7917766024 Dataset folder manager (#274)
* feat: retry send

* perf: qa default value

* feat: dataset folder

* feat: kb folder delete and path

* fix: ts

* perf: script load

* feat: fileCard and dataCard

* feat: search file

* feat: max token

* feat: select dataset

* fix: preview chunk

* perf: source update

* export data limit file_id

* docs

* fix: export limit
2023-09-10 16:37:32 +08:00
不做了睡大觉
a1a63260dd 更新镜像通道 (#272)
* 更新镜像

* 更新镜像信息

* 更新镜像信息
2023-09-08 18:13:37 +08:00
Archer
6f2d556a87 demo (#266) 2023-09-06 18:56:59 +08:00
Carson Yang
565f9c8113 Docs: fix typo (#265)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-06 14:29:49 +08:00
Carson Yang
975e011e03 Docs: update table style (#264)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-06 13:54:23 +08:00
Carson Yang
19ce6f66ca Docs: fix typo (#263)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-06 13:39:47 +08:00
cuisongliu
da6e26f95c build(main): add docs for ci (#261)
Signed-off-by: cuisongliu <cuisongliu@qq.com>
2023-09-06 12:06:51 +08:00
archer
71abe08f05 fix: onwechat yml 2023-09-06 10:46:55 +08:00
Carson Yang
45ba5e1e01 Docs: optimize codeblock (#259)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-06 09:53:37 +08:00
Carson Yang
139d0be52b Docs: fix favicon (#258)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-05 21:28:13 +08:00
Carson Yang
1ba3d72a8a Update docs: change image width (#256)
Signed-off-by: Carson Yang <yangchuansheng33@gmail.com>
2023-09-05 20:51:50 +08:00
archer
cd455b2a79 prompt docs 2023-09-05 20:22:22 +08:00
archer
fa3f3e6264 limit prompt template 2023-09-05 18:29:18 +08:00
archer
9bf5a3ec76 m3e doc 2023-09-05 18:29:15 +08:00
qing hua
95389e31f7 修改了一些错误 (#254)
* Update sse.ts

解决chatglm2控制台输出一半不输出的问题

* 解决知识库相似度搜索结果超过1

* Update sse.ts

---------

Co-authored-by: Archer <545436317@qq.com>
2023-09-05 18:28:59 +08:00
archer
ea65d9b34b onwechat demo 2023-09-05 14:31:04 +08:00
archer
2dd2976efa perf: dev doc 2023-09-05 11:47:04 +08:00
archer
64fde42c87 perf: default lang 2023-09-05 11:44:31 +08:00
archer
7a926b7086 user timezone 2023-09-05 11:30:52 +08:00
archer
562fd2692d issue template 2023-09-04 23:44:08 +08:00
archer
935287a95a doc favicon 2023-09-04 19:45:24 +08:00
archer
bd419a22f4 adapt echarts 2023-09-04 19:25:28 +08:00
不做了睡大觉
32f482b232 增加对echarts图表的支持 (#249)
* 增加对echarts图表的支持

* 增加对echarts的支持
2023-09-04 18:17:39 +08:00
archer
5d596bd3d5 favicon html 2023-09-04 18:16:29 +08:00
archer
ae88d79d6f update bash 2023-09-04 18:03:20 +08:00
archer
1207e3e566 update bash 2023-09-04 17:51:56 +08:00
archer
3449024678 feat: labBot demo 2023-09-04 17:02:21 +08:00
archer
8dba2c39e1 fix: empty kb 2023-09-04 15:47:01 +08:00
archer
94c53804ce fix: quick question and variable 2023-09-04 14:45:01 +08:00
archer
a1bcd798e1 image name 2023-09-04 14:31:03 +08:00
archer
6d51b3babe README 2023-09-04 11:37:46 +08:00
archer
a5fe671ffe perf: timeout 2023-09-04 11:32:49 +08:00
archer
44e772f0fd feat: file relate kb 2023-09-04 10:51:57 +08:00
archer
a3c6d6800b fix: file id 2023-09-03 22:59:11 +08:00
archer
19b1ff5a8d version docs 2023-09-03 22:52:17 +08:00
archer
8d55587cf4 version docs 2023-09-03 22:43:24 +08:00
archer
a754ceaf3b dataset save raw file 2023-09-03 22:39:09 +08:00
archer
086ea83fac docs 2023-09-03 19:06:12 +08:00
archer
1ace8fb9a3 error track 2023-09-03 18:26:36 +08:00
archer
e0b23a26f2 feat: error track, app scroll 2023-09-03 17:43:15 +08:00
archer
7c16d08ec0 feat: crud file 2023-09-03 17:43:14 +08:00
archer
5157e62fed feat: gridfs save file 2023-09-03 17:43:13 +08:00
archer
1fe2c49204 fastgpt char 2023-09-02 00:28:08 +08:00
archer
b9b50a0f5a README 2023-09-01 23:57:58 +08:00
archer
23cc2f81e9 perf: code 2023-09-01 11:47:06 +08:00
archer
68cdf50cb6 perf: config home page 2023-09-01 10:22:31 +08:00
archer
2ae8d43216 fix: reg failed 2023-08-31 20:06:49 +08:00
archer
0ea464f30f docs 2023-08-31 19:04:56 +08:00
archer
7cb035ba24 docs 2023-08-31 18:54:41 +08:00
archer
7231a847f7 fix: variable 2023-08-31 18:46:48 +08:00
archer
4f0f950dd4 docs 2023-08-31 18:43:10 +08:00
archer
c1f4785392 docs 2023-08-31 18:16:24 +08:00
archer
b22c878cf9 fix: variable input and update chat time 2023-08-31 18:09:33 +08:00
archer
3420f677b6 env template 2023-08-31 18:09:32 +08:00
不做了睡大觉
baee8cfe82 私有化模型对接oneapi教程+镜像更新 (#237)
* chatglm2-m3e对接教程

* chatglm2docker部署+对接Oneapi

* Update m3e.md
2023-08-31 17:58:48 +08:00
archer
0b0570fa54 perf: mark icon show 2023-08-30 14:31:46 +08:00
gaord
299409aa7b add python requirement.txt for running GLM2 model (#226)
Signed-off-by: Ben Gao <bengao168@msn.com>
2023-08-30 08:18:03 +08:00
archer
5284312eb3 chat default model 2023-08-29 18:22:45 +08:00
archer
86a0e7ce23 perf: guide modules 2023-08-29 17:59:24 +08:00
archer
e0de04dddb perf: open push data api 2023-08-29 11:07:25 +08:00
archer
19d7edb585 readme 2023-08-28 22:09:42 +08:00
archer
fbb75c97d0 docs 2023-08-28 21:59:25 +08:00
archer
7e9cac3478 feat: config login tip 2023-08-28 21:43:15 +08:00
archer
be937956af perf: config home title 2023-08-28 21:36:37 +08:00
archer
c5c3826714 perf: vector over range error 2023-08-28 15:51:28 +08:00
archer
42fec3a95c perf: git login 2023-08-28 15:30:18 +08:00
archer
64b9367ca1 perf: quote output prompt 2023-08-28 13:47:33 +08:00
943 changed files with 43270 additions and 51194 deletions

View File

@@ -4,21 +4,22 @@ about: 详细清晰的描述你遇到的问题
title: ''
labels: bug
assignees: ''
---
**例行检查**
[//]: # (方框内删除已有的空格,填 x 号)
+ [ ] 我已确认目前没有类似 issue
+ [ ] 我已完整查看过项目 README以及[项目文档](https://doc.fastgpt.run/docs/intro/)
+ [ ] 我使用了自己的key并确认我的 key 是可正常使用的
+ [ ] 我理解并愿意跟进此 issue协助测试和提供反馈
+ [ ] 我理解并认可上述内容,并理解项目维护者精力有限,**不遵循规则的 issue 可能会被无视或直接关闭**
[//]: # '方框内填 x 表示打钩'
- [ ] 我已确认目前没有类似 issue
- [ ]已完整查看过项目 README以及[项目文档](https://doc.fastgpt.run/docs/intro/)
- [ ]使用了自己的 key并确认我的 key 是可正常使用的
- [ ] 我理解并愿意跟进此 issue协助测试和提供反馈
- [x] 我理解并认可上述内容,并理解项目维护者精力有限,**不遵循规则的 issue 可能会被无视或直接关闭**
**你的版本**
+ [ ] 公有云版本
+ [ ] 私有部署版本
- [ ] 公有云版本
- [ ] 私有部署版本
**问题描述**

View File

@@ -8,13 +8,13 @@ assignees: ''
**例行检查**
[//]: # '方框内删除已有的空格,填 x 号'
[//]: # '方框内填 x 表示打钩'
- [ ] 我已确认目前没有类似 features
- [ ] 我已确认我已升级到最新版本
- [ ] 我已完整查看过项目 README已确定现有版本无法满足需求
- [ ] 我理解并愿意跟进此 features协助测试和提供反馈
- [ ] 我理解并认可上述内容,并理解项目维护者精力有限,**不遵循规则的 features 可能会被无视或直接关闭**
- [x] 我理解并认可上述内容,并理解项目维护者精力有限,**不遵循规则的 features 可能会被无视或直接关闭**
**功能描述**

30
.github/gh-bot.yml vendored Normal file
View File

@@ -0,0 +1,30 @@
version: v1
debug: true
action:
printConfig: false
release:
retry: 15s
actionName: Release
allowOps:
- cuisongliu
bot:
prefix: /
spe: _
allowOps:
- sealos-ci-robot
- sealos-release-robot
email: sealos-ci-robot@sealos.io
username: sealos-ci-robot
repo:
org: false
message:
success: |
🤖 says: Hooray! The action {{.Body}} has been completed successfully. 🎉
format_error: |
🤖 says: ‼️ There is a formatting issue with the action, kindly verify the action's format.
permission_error: |
🤖 says: ‼️ The action doesn't have permission to trigger.
release_error: |
🤖 says: ‼️ Release action failed.
Error details: {{.Error}}

15
.github/imgs/logo-left.svg vendored Normal file

File diff suppressed because one or more lines are too long

After

Width:  |  Height:  |  Size: 10 KiB

View File

@@ -0,0 +1,19 @@
name: 'Github Rebot for issues-translator'
on:
issues:
types: [ opened ]
issue_comment:
types: [ created ]
jobs:
translate:
permissions:
issues: write
discussions: write
pull-requests: write
runs-on: ubuntu-latest
steps:
- uses: usthe/issues-translate-action@v2.7
with:
IS_MODIFY_TITLE: true
BOT_GITHUB_TOKEN: ${{ secrets.GH_PAT }}
CUSTOM_BOT_NOTE: Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

View File

@@ -55,8 +55,6 @@ jobs:
# Step 4 - Builds the site using Hugo
- name: Build
run: cd docSite && hugo mod get -u github.com/colinwilson/lotusdocs && hugo -v --minify
env:
HUGO_BASEURL: ${{ vars.BASE_URL }}
# Step 5 - Push our generated site to Vercel
- name: Deploy to Vercel
@@ -69,3 +67,4 @@ jobs:
github-comment: false
vercel-args: '--prod --local-config ../vercel.json' # Optional
working-directory: docSite/public

94
.github/workflows/deploy-preview.yml vendored Normal file
View File

@@ -0,0 +1,94 @@
name: deploy-docs-preview
on:
pull_request_target:
paths:
- 'docSite/**'
branches:
- 'main'
workflow_dispatch:
# A workflow run is made up of one or more jobs that can run sequentially or in parallel
jobs:
# This workflow contains jobs "deploy-production"
deploy-preview:
# The environment this job references
environment:
name: Preview
url: ${{ steps.vercel-action.outputs.preview-url }}
# The type of runner that the job will run on
runs-on: ubuntu-22.04
# Job outputs
outputs:
url: ${{ steps.vercel-action.outputs.preview-url }}
# Steps represent a sequence of tasks that will be executed as part of the job
steps:
# Step 1 - Checks-out your repository under $GITHUB_WORKSPACE
- name: Checkout
uses: actions/checkout@v3
with:
ref: ${{ github.event.pull_request.head.ref }}
repository: ${{ github.event.pull_request.head.repo.full_name }}
submodules: recursive # Fetch submodules
fetch-depth: 0 # Fetch all history for .GitInfo and .Lastmod
# Step 2 Detect changes to Docs Content
- name: Detect changes in doc content
uses: dorny/paths-filter@v2
id: filter
with:
filters: |
docs:
- 'docSite/content/docs/**'
base: main
# Step 3 - Install Hugo (specific version)
- name: Install Hugo
uses: peaceiris/actions-hugo@v2
with:
hugo-version: '0.117.0'
extended: true
# Step 4 - Builds the site using Hugo
- name: Build
run: cd docSite && hugo mod get -u github.com/colinwilson/lotusdocs && hugo -v --minify
# Step 5 - Push our generated site to Vercel
- name: Deploy to Vercel
uses: amondnet/vercel-action@v25
id: vercel-action
with:
vercel-token: ${{ secrets.VERCEL_TOKEN }} # Required
vercel-org-id: ${{ secrets.VERCEL_ORG_ID }} #Required
vercel-project-id: ${{ secrets.VERCEL_PROJECT_ID }} #Required
github-comment: false
vercel-args: '--local-config ../vercel.json' # Optional
working-directory: docSite/public
alias-domains: | #Optional
fastgpt-staging.vercel.app
docsOutput:
needs: [ deploy-preview ]
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
with:
ref: ${{ github.event.pull_request.head.ref }}
repository: ${{ github.event.pull_request.head.repo.full_name }}
- name: Write md
run: |
echo "# 🤖 Generated by deploy action" > report.md
echo "[👀 Visit Preview](${{ needs.deploy-preview.outputs.url }})" >> report.md
cat report.md
- name: Gh Rebot for Sealos
uses: labring/gh-rebot@v0.0.6
if: ${{ (github.event_name == 'pull_request_target') }}
with:
version: v0.0.6
env:
GH_TOKEN: "${{ secrets.GH_PAT }}"
SEALOS_TYPE: "pr_comment"
SEALOS_FILENAME: "report.md"
SEALOS_REPLACE_TAG: "DEFAULT_REPLACE_DEPLOY"

98
.github/workflows/docs-image.yml vendored Normal file
View File

@@ -0,0 +1,98 @@
name: Build FastGPT docs images and copy image to docker hub
on:
workflow_dispatch:
push:
paths:
- 'docSite/**'
branches:
- 'main'
tags:
- 'v*.*.*'
jobs:
build-fastgpt-docs-images:
runs-on: ubuntu-20.04
steps:
- name: Checkout
uses: actions/checkout@v3
with:
fetch-depth: 1
- name: Set up QEMU (optional)
uses: docker/setup-qemu-action@v2
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v2
with:
driver-opts: network=host
- name: Cache Docker layers
uses: actions/cache@v2
with:
path: /tmp/.buildx-cache
key: ${{ runner.os }}-buildx-${{ github.sha }}
restore-keys: |
${{ runner.os }}-buildx-
- name: Login to GitHub Container Registry
uses: docker/login-action@v2
with:
registry: ghcr.io
username: ${{ github.repository_owner }}
password: ${{ secrets.GH_PAT }}
- name: Set DOCKER_REPO_TAGGED based on branch or tag
run: |
if [[ "${{ github.ref_name }}" == "main" ]]; then
echo "DOCKER_REPO_TAGGED=ghcr.io/${{ github.repository_owner }}/fastgpt-docs:latest" >> $GITHUB_ENV
else
echo "DOCKER_REPO_TAGGED=ghcr.io/${{ github.repository_owner }}/fastgpt-docs:${{ github.ref_name }}" >> $GITHUB_ENV
fi
- name: Build and publish image for main branch or tag push event
env:
DOCKER_REPO_TAGGED: ${{ env.DOCKER_REPO_TAGGED }}
run: |
docker buildx build \
--build-arg name=app \
--platform linux/amd64,linux/arm64 \
--label "org.opencontainers.image.source= https://github.com/ ${{ github.repository_owner }}/FastGPT" \
--label "org.opencontainers.image.description=fastgpt image" \
--label "org.opencontainers.image.licenses=Apache" \
--push \
--cache-from=type=local,src=/tmp/.buildx-cache \
--cache-to=type=local,dest=/tmp/.buildx-cache \
-t ${DOCKER_REPO_TAGGED} \
-f docSite/Dockerfile \
.
push-to-docker-hub:
needs: build-fastgpt-docs-images
runs-on: ubuntu-20.04
if: github.repository == 'labring/FastGPT'
steps:
- name: Checkout code
uses: actions/checkout@v3
- name: Login to Docker Hub
uses: docker/login-action@v2
with:
username: ${{ secrets.DOCKER_HUB_NAME }}
password: ${{ secrets.DOCKER_HUB_PASSWORD }}
- name: Set DOCKER_REPO_TAGGED based on branch or tag
run: |
if [[ "${{ github.ref_name }}" == "main" ]]; then
echo "IMAGE_TAG=latest" >> $GITHUB_ENV
else
echo "IMAGE_TAG=${{ github.ref_name }}" >> $GITHUB_ENV
fi
- name: Pull image from GitHub Container Registry
run: docker pull ghcr.io/${{ github.repository_owner }}/fastgpt-docs:${{env.IMAGE_TAG}}
- name: Tag image with Docker Hub repository name and version tag
run: docker tag ghcr.io/${{ github.repository_owner }}/fastgpt-docs:${{env.IMAGE_TAG}} ${{ secrets.DOCKER_IMAGE_NAME }}:${{env.IMAGE_TAG}}
- name: Push image to Docker Hub
run: docker push ${{ secrets.DOCKER_IMAGE_NAME }}:${{env.IMAGE_TAG}}
update-docs-image:
needs: build-fastgpt-docs-images
runs-on: ubuntu-20.04
if: github.repository == 'labring/FastGPT'
steps:
- name: Checkout code
uses: actions/checkout@v3
- uses: actions-hub/kubectl@master
env:
KUBE_CONFIG: ${{ secrets.KUBE_CONFIG }}
with:
args: rollout restart deployment fastgpt-docs

View File

@@ -1,9 +1,10 @@
name: Build fastgpt images and copy image to docker hub
name: Build FastGPT images and copy image to docker hub
on:
workflow_dispatch:
push:
paths:
- 'client/**'
- 'projects/app/**'
- 'packages/**'
branches:
- 'main'
tags:
@@ -25,6 +26,13 @@ jobs:
uses: docker/setup-buildx-action@v2
with:
driver-opts: network=host
- name: Cache Docker layers
uses: actions/cache@v2
with:
path: /tmp/.buildx-cache
key: ${{ runner.os }}-buildx-${{ github.sha }}
restore-keys: |
${{ runner.os }}-buildx-
- name: Login to GitHub Container Registry
uses: docker/login-action@v2
with:
@@ -38,24 +46,26 @@ jobs:
else
echo "DOCKER_REPO_TAGGED=ghcr.io/${{ github.repository_owner }}/fastgpt:${{ github.ref_name }}" >> $GITHUB_ENV
fi
- name: Build and publish image for main branch or tag push event
env:
DOCKER_REPO_TAGGED: ${{ env.DOCKER_REPO_TAGGED }}
run: |
cd client && \
docker buildx build \
--build-arg name=app \
--platform linux/amd64,linux/arm64 \
--label "org.opencontainers.image.source= https://github.com/ ${{ github.repository_owner }}/FastGPT" \
--label "org.opencontainers.image.description=fastgpt image" \
--label "org.opencontainers.image.licenses=MIT" \
--label "org.opencontainers.image.licenses=Apache" \
--push \
--cache-from=type=local,src=/tmp/.buildx-cache \
--cache-to=type=local,dest=/tmp/.buildx-cache \
-t ${DOCKER_REPO_TAGGED} \
-f Dockerfile \
.
push-to-docker-hub:
needs: build-fastgpt-images
runs-on: ubuntu-20.04
if: github.repository == 'labring/FastGPT'
steps:
- name: Checkout code
uses: actions/checkout@v3
@@ -79,6 +89,7 @@ jobs:
run: docker push ${{ secrets.DOCKER_IMAGE_NAME }}:${{env.IMAGE_TAG}}
push-to-ali-hub:
needs: build-fastgpt-images
if: github.repository == 'labring/FastGPT'
runs-on: ubuntu-20.04
steps:
- name: Checkout code

55
.github/workflows/preview-image.yml vendored Normal file
View File

@@ -0,0 +1,55 @@
name: Preview FastGPT images
on:
pull_request_target:
paths:
- 'projects/app/**'
- 'packages/**'
branches:
- 'main'
workflow_dispatch:
jobs:
build-fastgpt-images:
runs-on: ubuntu-20.04
steps:
- name: Checkout
uses: actions/checkout@v3
with:
ref: ${{ github.event.pull_request.head.ref }}
repository: ${{ github.event.pull_request.head.repo.full_name }}
submodules: recursive # Fetch submodules
fetch-depth: 0 # Fetch all history for .GitInfo and .Lastmod
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v2
with:
driver-opts: network=host
- name: Cache Docker layers
uses: actions/cache@v2
with:
path: /tmp/.buildx-cache
key: ${{ runner.os }}-buildx-${{ github.sha }}
restore-keys: |
${{ runner.os }}-buildx-
- name: Login to GitHub Container Registry
uses: docker/login-action@v2
with:
registry: ghcr.io
username: ${{ github.repository_owner }}
password: ${{ secrets.GH_PAT }}
- name: Set DOCKER_REPO_TAGGED based on branch or tag
run: |
echo "DOCKER_REPO_TAGGED=ghcr.io/${{ github.repository_owner }}/fastgpt-pr:${{ github.event.pull_request.number }}" >> $GITHUB_ENV
- name: Build image for PR
env:
DOCKER_REPO_TAGGED: ${{ env.DOCKER_REPO_TAGGED }}
run: |
docker buildx build \
--build-arg name=app \
--label "org.opencontainers.image.source= https://github.com/ ${{ github.repository_owner }}/FastGPT" \
--label "org.opencontainers.image.description=fastgpt-pr image" \
--label "org.opencontainers.image.licenses=Apache" \
--cache-from=type=local,src=/tmp/.buildx-cache \
--cache-to=type=local,dest=/tmp/.buildx-cache \
-t ${DOCKER_REPO_TAGGED} \
-f Dockerfile \
.

4
.gitignore vendored
View File

@@ -33,4 +33,6 @@ dist/
# hugo
**/.hugo_build.lock
docSite/public/
docSite/public/
docSite/resources/_gen/
docSite/.vercel

View File

@@ -1,4 +1,5 @@
dist
.vscode
**/.DS_Store
node_modules
node_modules
docSite/

View File

@@ -1,10 +1,10 @@
{
"editor.formatOnSave": true,
"editor.mouseWheelZoom": true,
"typescript.tsdk": "client/node_modules/typescript/lib",
"typescript.tsdk": "node_modules/typescript/lib",
"prettier.prettierPath": "./node_modules/prettier",
"i18n-ally.localesPaths": [
"client/public/locales"
"projects/app/public/locales"
],
"i18n-ally.enabledParsers": ["json"],
"i18n-ally.keystyle": "nested",

68
Dockerfile Normal file
View File

@@ -0,0 +1,68 @@
# Install dependencies only when needed
FROM node:18.15-alpine AS deps
# Check https://github.com/nodejs/docker-node/tree/b4117f9333da4138b03a546ec926ef50a31506c3#nodealpine to understand why libc6-compat might be needed.
RUN apk add --no-cache libc6-compat && npm install -g pnpm
WORKDIR /app
ARG name
# copy packages and one project
COPY package.json pnpm-lock.yaml pnpm-workspace.yaml ./
COPY ./packages ./packages
COPY ./projects/$name/package.json ./projects/$name/package.json
RUN [ -f pnpm-lock.yaml ] || (echo "Lockfile not found." && exit 1)
RUN pnpm install
# Rebuild the source code only when needed
FROM node:18.15-alpine AS builder
WORKDIR /app
ARG name
# copy common node_modules and one project node_modules
COPY package.json pnpm-workspace.yaml ./
COPY --from=deps /app/node_modules ./node_modules
COPY --from=deps /app/packages ./packages
COPY ./projects/$name ./projects/$name
COPY --from=deps /app/projects/$name/node_modules ./projects/$name/node_modules
# Uncomment the following line in case you want to disable telemetry during the build.
ENV NEXT_TELEMETRY_DISABLED 1
RUN npm install -g pnpm
RUN pnpm --filter=$name run build
FROM node:18.15-alpine AS runner
WORKDIR /app
ARG name
# create user and use it
RUN addgroup --system --gid 1001 nodejs
RUN adduser --system --uid 1001 nextjs
RUN sed -i 's/https/http/' /etc/apk/repositories
RUN apk add curl \
&& apk add ca-certificates \
&& update-ca-certificates
# copy running files
COPY --from=builder /app/projects/$name/public ./projects/$name/public
COPY --from=builder /app/projects/$name/next.config.js ./projects/$name/next.config.js
COPY --from=builder --chown=nextjs:nodejs /app/projects/$name/.next/standalone ./
COPY --from=builder --chown=nextjs:nodejs /app/projects/$name/.next/static ./projects/$name/.next/static
# copy package.json to version file
COPY --from=builder /app/projects/$name/package.json ./package.json
ENV NODE_ENV production
ENV NEXT_TELEMETRY_DISABLED 1
ENV PORT=3000
EXPOSE 3000
USER nextjs
ENV serverPath=./projects/$name/server.js
ENTRYPOINT ["sh","-c","node ${serverPath}"]

110
README.md
View File

@@ -4,23 +4,39 @@
# FastGPT
<p align="center">
<a href="./README_en.md">English</a> |
<a href="./README.md">简体中文</a>
</p>
FastGPT 是一个基于 LLM 大语言模型的知识库问答系统,提供开箱即用的数据处理、模型调用等能力。同时可以通过 Flow 可视化进行工作流编排,从而实现复杂的问答场景!
</div>
<p align="center">
<a href="https://fastgpt.run/">线上体验</a>
·
<a href="https://doc.fastgpt.run/docs/intro">相关文档</a>
·
<a href="https://doc.fastgpt.run/docs/development">本地开发</a>
·
<a href="https://github.com/labring/FastGPT#-%E7%9B%B8%E5%85%B3%E9%A1%B9%E7%9B%AE">相关项目</a>
<a href="https://fastgpt.run/">
<img height="21" src="https://img.shields.io/badge/在线使用-d4eaf7?style=flat-square&logo=spoj&logoColor=7d09f1" alt="cloud">
</a>
<a href="https://doc.fastgpt.run/docs/intro">
<img height="21" src="https://img.shields.io/badge/相关文档-7d09f1?style=flat-square" alt="document">
</a>
<a href="https://doc.fastgpt.run/docs/development">
<img height="21" src="https://img.shields.io/badge/本地开发-%23d4eaf7?style=flat-square&logo=xcode&logoColor=7d09f1" alt="development">
</a>
<a href="/#-%E7%9B%B8%E5%85%B3%E9%A1%B9%E7%9B%AE">
<img height="21" src="https://img.shields.io/badge/相关项目-7d09f1?style=flat-square" alt="project">
</a>
<a href="https://github.com/labring/FastGPT/blob/main/LICENSE">
<img height="21" src="https://img.shields.io/badge/License-Apache--2.0-ffffff?style=flat-square&labelColor=d4eaf7&color=7d09f1" alt="license">
</a>
</p>
## 🛸 在线体验
https://github.com/labring/FastGPT/assets/15308462/7d3a38df-eb0e-4388-9250-2409bd33f6d4
[fastgpt.run](https://fastgpt.run/)(服务器在新加坡,部分地区可能无法直连)
## 🛸 在线使用
- 🌐 国内版:[ai.fastgpt.in](https://ai.fastgpt.in/)
- 🌍 海外版:[fastgpt.run](https://fastgpt.run/)
| | |
| ---------------------------------- | ---------------------------------- |
@@ -31,42 +47,40 @@ FastGPT 是一个基于 LLM 大语言模型的知识库问答系统,提供开
1. 强大的可视化编排,轻松构建 AI 应用
- [x] 提供简易模式,无需操作编排
- [x] 用户对话前引导
- [x] 全局变量
- [x] 用户对话前引导,全局字符串变量
- [x] 知识库搜索
- [x] 多 LLM 模型对话
- [x] 文本内容提取成结构化数据
- [x] HTTP 扩展
- [ ] 沙盒 JS 运行模块
- [ ] 连续对话引导
- [ ] 嵌入 Laf实现在线编写 HTTP 模块
- [x] 对话下一步指引
- [ ] 对话多路线选择
- [ ] 源文件引用追踪
- [x] 源文件引用追踪
- [x] 模块封装,实现多级复用
2. 丰富的知识库预处理
- [x] 多库复用,混用
- [x] chunk 记录修改和删除
- [x] 支持直接分段导入
- [x] 支持 QA 拆分导入
- [x] 支持手动输入内容
- [x] 支持 url 读取导入
- [x] 支持 CSV 批量导入问答对
- [ ] 支持知识库单独设置向量模型
- [ ] 源文件存储
- [x] 支持手动输入直接分段QA 拆分导入
- [x] 支持 url 读取、CSV 批量导入
- [x] 支持知识库单独设置向量模型
- [x] 源文件存储
- [ ] 文件学习 Agent
3. 多种效果测试渠道
- [x] 知识库单点搜索测试
- [x] 对话时反馈引用并可修改与删除
- [x] 完整上下文呈现
- [ ] 完整模块中间值呈现
- [x] 完整模块中间值呈现
4. OpenAPI
- [x] completions 接口对齐 GPT 接口
- [x] completions 接口 (对齐 GPT 接口)
- [ ] 知识库 CRUD
5. 运营功能
- [x] 免登录分享窗口
- [x] Iframe 一键嵌入
- [ ] 统一查阅对话记录
- [x] 统一查阅对话记录,并对数据进行标注
## 👨‍💻 开发
项目技术栈: NextJs + TS + ChakraUI + Mongo + PostgresVector 插件
项目技术栈NextJs + TS + ChakraUI + Mongo + Postgres (Vector 插件)
- **⚡ 快速部署**
@@ -76,36 +90,36 @@ FastGPT 是一个基于 LLM 大语言模型的知识库问答系统,提供开
由于需要部署数据库,部署完后需要等待 2~4 分钟才能正常访问。默认用了最低配置,首次访问时会有些慢。
* [快开始本地开发](https://doc.fastgpt.run/docs/development)
* [部署 FastGPT](https://doc.fastgpt.run/docs/installation)
* [系统配置文件说明](https://doc.fastgpt.run/docs/installation/reference)
* [多模型配置](https://doc.fastgpt.run/docs/installation/reference/models)
* [版本升级](https://doc.fastgpt.run/docs/installation/upgrading)
* [API 文档](https://kjqvjse66l.feishu.cn/docx/DmLedTWtUoNGX8xui9ocdUEjnNh?pre_pathname=%2Fdrive%2Fhome%2F)
* [快开始本地开发](https://doc.fastgpt.in/docs/development/intro/)
* [部署 FastGPT](https://doc.fastgpt.in/docs/installation)
* [系统配置文件说明](https://doc.fastgpt.in/docs/development/configuration/)
* [多模型配置](https://doc.fastgpt.in/docs/installation/one-api/)
* [版本更新/升级介绍](https://doc.fastgpt.in/docs/installation/upgrading)
* [OpenAPI API 文档](https://doc.fastgpt.in/docs/development/openapi/)
* [知识库结构详解](https://doc.fastgpt.in/docs/use-cases/datasetengine/)
## 🏘️ 社区交流群
| 交流群 | 小助手 |
| ----------------------------------------------------- | ---------------------------------------------- |
| ![](https://otnvvf-imgs.oss.laf.run/wxqun300-222.jpg) | ![](https://otnvvf-imgs.oss.laf.run/wx300.jpg) |
添加 wx 小助手加入:
## 👀 其他
- [FastGpt 常见问题](https://kjqvjse66l.feishu.cn/docx/HtrgdT0pkonP4kxGx8qcu6XDnGh)
- [docker 部署教程视频](https://www.bilibili.com/video/BV1jo4y147fT/)
- [公众号接入视频教程](https://www.bilibili.com/video/BV1xh4y1t7fy/)
- [FastGpt 知识库演示](https://www.bilibili.com/video/BV1Wo4y1p7i1/)
![](https://otnvvf-imgs.oss.laf.run/wx300.jpg)
## 💪 相关项目
- [Laf: 3 分钟快速接入三方应用](https://github.com/labring/laf)
- [Sealos: 快速部署集群应用](https://github.com/labring/sealos)
- [One API: 多模型管理,支持 Azure、文心一言等](https://github.com/songquanpeng/one-api)
- [TuShan: 5 分钟搭建后台管理系统](https://github.com/msgbyte/tushan)
- [Laf3 分钟快速接入三方应用](https://github.com/labring/laf)
- [Sealos快速部署集群应用](https://github.com/labring/sealos)
- [One API多模型管理,支持 Azure、文心一言等](https://github.com/songquanpeng/one-api)
- [TuShan5 分钟搭建后台管理系统](https://github.com/msgbyte/tushan)
## 👀 其他
- [保姆级 FastGPT 教程](https://www.bilibili.com/video/BV1n34y1A7Bo/?spm_id_from=333.999.0.0)
- [接入飞书](https://www.bilibili.com/video/BV1Su4y1r7R3/?spm_id_from=333.999.0.0)
- [接入企微](https://www.bilibili.com/video/BV1Tp4y1n72T/?spm_id_from=333.999.0.0)
## 🤝 第三方生态
- [luolinAI: 企微机器人,开箱即用](https://github.com/luolin-ai/FastGPT-Enterprise-WeChatbot)
- [OnWeChat 个人微信/企微机器人](https://doc.fastgpt.run/docs/use-cases/onwechat/)
## 🌟 Star History
@@ -115,7 +129,7 @@ FastGPT 是一个基于 LLM 大语言模型的知识库问答系统,提供开
本仓库遵循 [FastGPT Open Source License](./LICENSE) 开源协议。
1. 允许作为后台服务直接商用,但不允许直接使用 saas 服务商用
2. 需保留相关版权信息。
1. 允许作为后台服务直接商用,但不允许提供 SaaS 服务。
2. 未经商业授权,任何形式的商用服务均需保留相关版权信息。
3. 完整请查看 [FastGPT Open Source License](./LICENSE)
4. 联系方式yujinlong@sealos.io, [点击查看定价策略](https://fael3z0zfze.feishu.cn/docx/F155dbirfo8vDDx2WgWc6extnwf)
4. 联系方式yujinlong@sealos.io[点击查看商业版定价策略](https://doc.fastgpt.run/docs/commercial)

View File

@@ -1,25 +1,39 @@
<div align="center">
<a href="https://fastgpt.run/"><img src="/.github/imgs/logo.svg" width="120" height="120" alt="fastgpt logo"></a>
# FastGPT
FastGPT is a knowledge-based question answering system based on the LLM language model, providing out-of-the-box capabilities for data processing, model invocation, and more. It also allows for complex question answering scenarios through visual workflow orchestration using Flow!
<p align="center">
<a href="./README_en.md">English</a> |
<a href="./README.md">简体中文</a>
</p>
FastGPT is a knowledge-based Q&A system built on the LLM, offers out-of-the-box data processing and model invocation capabilities, allows for workflow orchestration through Flow visualization!
</div>
<p align="center">
<a href="https://fastgpt.run/">Online</a>
·
<a href="https://doc.fastgpt.run/docs/intro">Document</a>
·
<a href="https://doc.fastgpt.run/docs/development">Development</a>
·
<a href="https://doc.fastgpt.run/docs/installation">Deploy</a>
·
<a href="#powered-by">Power By</a>
<a href="https://fastgpt.run/">
<img height="21" src="https://img.shields.io/badge/在线使用-d4eaf7?style=flat-square&logo=spoj&logoColor=7d09f1" alt="cloud">
</a>
<a href="https://doc.fastgpt.run/docs/intro">
<img height="21" src="https://img.shields.io/badge/相关文档-7d09f1?style=flat-square" alt="document">
</a>
<a href="https://doc.fastgpt.run/docs/development">
<img height="21" src="https://img.shields.io/badge/本地开发-%23d4eaf7?style=flat-square&logo=xcode&logoColor=7d09f1" alt="development">
</a>
<a href="/#-%E7%9B%B8%E5%85%B3%E9%A1%B9%E7%9B%AE">
<img height="21" src="https://img.shields.io/badge/相关项目-7d09f1?style=flat-square" alt="project">
</a>
<a href="https://github.com/labring/FastGPT/blob/main/LICENSE">
<img height="21" src="https://img.shields.io/badge/License-Apache--2.0-ffffff?style=flat-square&labelColor=d4eaf7&color=7d09f1" alt="license">
</a>
</p>
## 🛸 Online
https://github.com/labring/FastGPT/assets/15308462/7d3a38df-eb0e-4388-9250-2409bd33f6d4
## 🛸 Use Cloud Services
[fastgpt.run](https://fastgpt.run/)
| | |
@@ -29,35 +43,34 @@ FastGPT is a knowledge-based question answering system based on the LLM language
## 💡 Features
1. Powerful visual orchestration for easy AI application building
1. Powerful visual workflows: Effortlessly craft AI applications
- [x] Provides a simple mode without the need for orchestration operations
- [x] Simple mode on deck - no need for manual arrangement
- [x] User dialogue pre-guidance
- [x] Global variables
- [x] Knowledge base search
- [x] Multi-LLM model dialogue
- [x] Extraction of text content into structured data
- [x] HTTP extension
- [ ] Sandbox JS runtime module
- [ ] Continuous dialogue guidance
- [ ] Dialogue multi-path selection
- [ ] Source file reference tracking
- [x] Dialogue via multiple LLM models
- [x] Text magic - convert to structured data
- [x] Extend with HTTP
- [ ] Embed Laf for on-the-fly HTTP module crafting
- [x] Directions for the next dialogue steps
- [x] Tracking source file references
- [ ] Custom file reader
- [ ] Modules are packaged into plug-ins to achieve reuse
2. Rich knowledge base preprocessing
2. Extensive knowledge base preprocessing
- [x] Multiple library reuse and mixing
- [x] Chunk record modification and deletion
- [x] Supports direct segment import
- [x] Supports QA split import
- [x] Supports manual input content
- [ ] Supports URL import reading
- [x] Supports batch import of Q&A pairs in CSV format
- [ ] Supports separate vector model settings for knowledge bases
- [ ] Source file storage
- [x] Reuse and mix multiple knowledge bases
- [x] Track chunk modifications and deletions
- [x] Supports manual entries, direct segmentation, and QA split imports
- [x] Supports URL fetching and batch CSV imports
- [x] Supports Set unique vector models for knowledge bases
- [x] Store original files
- [ ] File learning Agent
3. Multiple effect testing channels
- [x] Knowledge base single point search testing
- [x] Single-point knowledge base search test
- [x] Feedback references and ability to modify and delete during dialogue
- [x] Complete context presentation
- [ ] Complete module intermediate value presentation
@@ -77,14 +90,20 @@ FastGPT is a knowledge-based question answering system based on the LLM language
Project tech stack: NextJs + TS + ChakraUI + Mongo + Postgres (Vector plugin)
- **⚡ Deployment**
[![](https://cdn.jsdelivr.us/gh/labring-actions/templates@main/Deploy-on-Sealos.svg)](https://cloud.sealos.io/?openapp=system-fastdeploy%3FtemplateName%3Dfastgpt)
Give it a 2-4 minute wait after deployment as it sets up the database. Initially, it might be a tad slow since we're using the basic settings.
- [Getting Started with Local Development](https://doc.fastgpt.run/docs/development)
- [Deploying FastGPT](https://doc.fastgpt.run/docs/installation)
- [System Configuration File Explanation](https://doc.fastgpt.run/docs/installation/reference)
- [Multi-model Configuration](https://doc.fastgpt.run/docs/installation/reference/models)
- [V3 Upgrade V4 Initialization](https://doc.fastgpt.run/docs/installation/upgrading)
- [Guide on System Configs](https://doc.fastgpt.run/docs/installation/reference)
- [Configuring Multiple Models](https://doc.fastgpt.run/docs/installation/reference/models)
- [Version Updates & Upgrades](https://doc.fastgpt.run/docs/installation/upgrading)
<!-- ## :point_right: RoadMap
- [FastGpt RoadMap](https://kjqvjse66l.feishu.cn/docx/RVUxdqE2WolDYyxEKATcM0XXnte) -->
- [FastGPT RoadMap](https://kjqvjse66l.feishu.cn/docx/RVUxdqE2WolDYyxEKATcM0XXnte) -->
<!-- ## 🏘️ Community
@@ -94,10 +113,10 @@ Project tech stack: NextJs + TS + ChakraUI + Mongo + Postgres (Vector plugin)
## 👀 Others
- [FastGpt FAQ](https://kjqvjse66l.feishu.cn/docx/HtrgdT0pkonP4kxGx8qcu6XDnGh)
- [FastGPT FAQ](https://kjqvjse66l.feishu.cn/docx/HtrgdT0pkonP4kxGx8qcu6XDnGh)
- [Docker Deployment Tutorial Video](https://www.bilibili.com/video/BV1jo4y147fT/)
- [Official Account Integration Video Tutorial](https://www.bilibili.com/video/BV1xh4y1t7fy/)
- [FastGpt Knowledge Base Demo](https://www.bilibili.com/video/BV1Wo4y1p7i1/)
- [FastGPT Knowledge Base Demo](https://www.bilibili.com/video/BV1Wo4y1p7i1/)
## 💪 Related Projects

View File

@@ -1,65 +0,0 @@
# Install dependencies only when needed
FROM node:current-alpine AS deps
# Check https://github.com/nodejs/docker-node/tree/b4117f9333da4138b03a546ec926ef50a31506c3#nodealpine to understand why libc6-compat might be needed.
RUN apk add --no-cache libc6-compat && npm install -g pnpm
WORKDIR /app
# Install dependencies based on the preferred package manager
COPY package.json ./
COPY pnpm-lock.yaml* ./
RUN \
[ -f pnpm-lock.yaml ] && pnpm fetch || \
(echo "Lockfile not found." && exit 1)
# Rebuild the source code only when needed
FROM node:current-alpine AS builder
WORKDIR /app
COPY --from=deps /app/node_modules ./node_modules
COPY pnpm-lock.yaml* ./
COPY package.json ./
COPY . .
# Next.js collects completely anonymous telemetry data about general usage.
# Learn more here: https://nextjs.org/telemetry
# Uncomment the following line in case you want to disable telemetry during the build.
ENV NEXT_TELEMETRY_DISABLED 1
RUN npm install -g pnpm
RUN \
[ -f pnpm-lock.yaml ] && (pnpm --offline install && pnpm run build) || \
(echo "Lockfile not found." && exit 1)
# Production image, copy all the files and run next
FROM node:current-alpine AS runner
WORKDIR /app
ENV NODE_ENV production
# Uncomment the following line in case you want to disable telemetry during runtime.
ENV NEXT_TELEMETRY_DISABLED 1
RUN addgroup --system --gid 1001 nodejs
RUN adduser --system --uid 1001 nextjs
RUN sed -i 's/https/http/' /etc/apk/repositories
RUN apk add curl \
&& apk add ca-certificates \
&& update-ca-certificates
# You only need to copy next.config.js if you are NOT using the default configuration
# COPY --from=builder /app/next.config.js ./
COPY --from=builder /app/public ./public
COPY --from=builder /app/package.json ./package.json
# COPY --from=builder /app/.env* .
# Automatically leverage output traces to reduce image size
# https://nextjs.org/docs/advanced-features/output-file-tracing
COPY --from=builder --chown=nextjs:nodejs /app/.next/standalone ./
COPY --from=builder --chown=nextjs:nodejs /app/.next/static ./.next/static
USER nextjs
ENV PORT=3000
EXPOSE 3000
CMD ["node", "server.js"]

View File

@@ -1,63 +0,0 @@
{
"FeConfig": {
"show_emptyChat": true,
"show_register": false,
"show_appStore": false,
"show_userDetail": false,
"show_git": true,
"systemTitle": "FastGPT",
"authorText": "Made by FastGPT Team.",
"gitLoginKey": "",
"scripts": []
},
"SystemParams": {
"gitLoginSecret": "",
"vectorMaxProcess": 15,
"qaMaxProcess": 15,
"pgIvfflatProbe": 20
},
"ChatModels": [
{
"model": "gpt-3.5-turbo",
"name": "GPT35-4k",
"contextMaxToken": 4000,
"quoteMaxToken": 2000,
"maxTemperature": 1.2,
"price": 0,
"defaultSystem": ""
},
{
"model": "gpt-3.5-turbo-16k",
"name": "GPT35-16k",
"contextMaxToken": 16000,
"quoteMaxToken": 8000,
"maxTemperature": 1.2,
"price": 0,
"defaultSystem": ""
},
{
"model": "gpt-4",
"name": "GPT4-8k",
"contextMaxToken": 8000,
"quoteMaxToken": 4000,
"maxTemperature": 1.2,
"price": 0,
"defaultSystem": ""
}
],
"VectorModels": [
{
"model": "text-embedding-ada-002",
"name": "Embedding-2",
"price": 0,
"defaultToken": 500,
"maxToken": 3000
}
],
"QAModel": {
"model": "gpt-3.5-turbo-16k",
"name": "GPT35-16k",
"maxToken": 16000,
"price": 0
}
}

View File

@@ -1,5 +0,0 @@
/// <reference types="next" />
/// <reference types="next/image-types/global" />
// NOTE: This file should not be edited
// see https://nextjs.org/docs/basic-features/typescript for more information.

12308
client/pnpm-lock.yaml generated

File diff suppressed because it is too large Load Diff

View File

@@ -1,22 +0,0 @@
### 常见问题
- [**Git 地址**,点击查看项目地址](https://github.com/labring/FastGPT)
- [本地部署 FastGPT](https://doc.fastgpt.run/docs/installation)
- [API 文档](https://kjqvjse66l.feishu.cn/docx/DmLedTWtUoNGX8xui9ocdUEjnNh?pre_pathname=%2Fdrive%2Fhome%2F)
- **反馈问卷**: 如果你遇到任何使用问题或有期望的功能,可以[填写该问卷](https://www.wjx.cn/vm/rLIw1uD.aspx#)
- **问题文档**: [先看文档,再提问](https://kjqvjse66l.feishu.cn/docx/HtrgdT0pkonP4kxGx8qcu6XDnGh)
- [点击查看商业版文档](https://fael3z0zfze.feishu.cn/docx/F155dbirfo8vDDx2WgWc6extnwf)
**价格表**
| 计费项 | 价格: 元/ 1K tokens包含上下文|
| --- | --- |
| 知识库 - 索引 | 0.002 |
| FastAI4k - 对话 | 0.015 |
| FastAI16k - 对话 | 0.03 |
| FastAI-Plus - 对话 | 0.45 |
| 文件拆分 | 0.03 |
**其他问题**
| 交流群 | 小助手 |
| ----------------------- | -------------------- |
| ![](https://otnvvf-imgs.oss.laf.run/wxqun300.jpg) | ![](https://otnvvf-imgs.oss.laf.run/wx300.jpg) |

View File

@@ -1,7 +0,0 @@
### Fast GPT V4.2.2
1. **新增 - 用户反馈和管理员标注预期答案,以不断提高模型回复准确率。** 该功能为测试版,未来交互可能会有变化,欢迎大家提出宝贵意见。
2. 优化 - 知识库搜索提示词,更适配问答场景。
3. 新增 - 好友邀请链接,[点击查看](/account?currentTab=promotion)
4. 优化 - [使用文档](https://doc.fastgpt.run/docs/intro/)
5. [点击查看高级编排介绍文档](https://doc.fastgpt.run/docs/workflow)

Binary file not shown.

Before

Width:  |  Height:  |  Size: 11 KiB

View File

@@ -1 +0,0 @@
<?xml version="1.0" standalone="no"?><!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd"><svg t="1692418843591" class="icon" viewBox="0 0 1024 1024" version="1.1" xmlns="http://www.w3.org/2000/svg" p-id="4084" xmlns:xlink="http://www.w3.org/1999/xlink" width="64" height="64"><path d="M511.5 82c-236.6 0-429 192.4-429 429 0 236.5 192.5 429 429 429 236.6 0 429-192.4 429-429 0-236.5-192.4-429-429-429z m377.6 403.8H734.3c-4-139.9-41.4-259.9-97.5-331.9C776.5 203 879 332 889.1 485.8z m-402.8-349v349h-147c5.5-175.5 68.6-322.6 147-349z m0 399.4v349c-78.4-26.4-141.4-173.5-147-349h147z m50.5 349v-349h147c-5.6 175.5-68.6 322.6-147 349z m0-399.4v-349c78.4 26.4 141.4 173.5 147 349h-147zM386.3 153.9c-56.1 72-93.5 192-97.5 331.9H133.9C144.1 332 246.5 203 386.3 153.9zM133.9 536.2h154.8c4 139.9 41.4 259.9 97.5 331.9C246.5 819 144.1 690 133.9 536.2z m502.8 331.9c56.1-72 93.5-192 97.5-331.9H889C879 690 776.5 819 636.7 868.1z" fill="#5F9BEB" p-id="4085"></path></svg>

Before

Width:  |  Height:  |  Size: 1006 B

View File

@@ -1,191 +0,0 @@
{
"App": "App",
"Cancel": "No",
"Confirm": "Yes",
"Running": "Running",
"Select value is empty": "Select value is empty",
"UnKnow": "UnKnow",
"Warning": "Warning",
"app": {
"Advance App TestTip": "The current application is advanced editing mode \n. If you need to switch to [simple mode], please click the save button on the left",
"App Detail": "App Detail",
"Chat Logs Tips": "Logs record the app's online, shared, and API conversations",
"Chat logs": "Chat Logs",
"Confirm Del App Tip": "Confirm to delete the app and all its chats",
"Confirm Save App Tip": "The application may be in advanced orchestration mode, and the advanced orchestration configuration will be overwritten after saving, please confirm!",
"Connection is invalid": "Connecting is invalid",
"Connection type is different": "Connection type is different",
"Copy Module Config": "Copy config",
"Export Config Successful": "The configuration has been copied. Please check for important data",
"Export Configs": "Export Configs",
"Feedback Count": "User Feedback",
"Import Config": "Import Config",
"Import Config Failed": "Failed to import the configuration, please ensure that the configuration is normal!",
"Import Configs": "Import Configs",
"Input Field Settings": "Input Field Settings",
"Logs Empty": "Logs is empty",
"Logs Message Total": "Message Count",
"Logs Source": "Source",
"Logs Time": "Time",
"Logs Title": "Title",
"Mark Count": "Mark Count",
"My Apps": "My Apps",
"Output Field Settings": "Output Field Settings",
"Paste Config": "Paste Config"
},
"chat": {
"Admin Mark Content": "Corrected response",
"Complete Response": "Complete Response",
"Confirm to clear history": "Confirm to clear history?",
"Exit Chat": "Exit",
"Feedback Failed": "Feedback Failed",
"Feedback Mark": "Mark",
"Feedback Modal": "Feedback",
"Feedback Modal Tip": "Enter what you find unsatisfactory",
"Feedback Close": "Close Feedback",
"Feedback Submit": "Submit",
"Feedback Success": "Feedback Success",
"Feedback Update Failed": "Feedback Update Failed",
"History": "History",
"Mark": "Mark",
"Mark Description": "The annotation feature is currently in beta. \n\n After clicking Add annotation, you need to select a knowledge base in order to store annotation data. You can use this feature to quickly annotate questions and expected answers to guide the model to the next answer. At present, the annotation function, like other data in the knowledge base, is affected by the model, which does not mean that the annotation meets 100% expectations. The \n\n annotation data is only unidirectional synchronization with the knowledge base. If the knowledge base modifies the annotation data, the annotation data displayed in the log cannot be synchronized",
"Mark Description Title": "Mark Description",
"New Chat": "New Chat",
"Read Mark Description": "Read mark description",
"Read User Feedback": "Read user feedback",
"Select Mark Kb": "Select Dataset",
"Select Mark Kb Desc": "Select a dataset to store the expected answers",
"You need to a chat app": "You need to a chat app",
"logs": {
"api": "API",
"online": "Online Chat",
"share": "Share",
"test": "Test Chat "
}
},
"commom": {
"Password inconsistency": "Password inconsistency"
},
"common": {
"Add": "Add",
"Cancel": "Cancel",
"Collect": "Collect",
"Copy": "Copy",
"Copy Successful": "Copy Successful",
"Course": "",
"Delete": "Delete",
"Filed is repeat": "Filed is repeated",
"Filed is repeated": "",
"Input": "Input",
"Output": "Output",
"export": ""
},
"dataset": {
"Confirm to delete the data": "Confirm to delete the data?",
"Queue Desc": "This data refers to the current amount of training for the entire system. FastGPT uses queued training, and if you have too much data to train, you may need to wait for a while",
"System Data Queue": "Data Queue"
},
"file": {
"Click to download CSV template": "Click to download CSV template",
"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",
"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",
"max 10": "Max 10 files",
"select a document": "select a document",
"support": "support {{fileExtension}} file",
"upload error description": "Only upload multiple files or one folder at a time"
},
"home": {
"AI Assistant": "AI Assistant",
"AI Assistant Desc": "",
"Advanced Settings": "",
"Advanced Settings Desc": "",
"Choice Debug": "Convenient Debugging",
"Choice Debug Desc": "Search testing, reference modification, full conversation preview and many other debugging ways",
"Choice Desc": "FastGPT follows the Apache License 2.0 open source protocol",
"Choice Extension": "Infinite Extension",
"Choice Extension Desc": "HTTP based extension, easy to achieve custom functions",
"Choice Models": "Multiple Models",
"Choice Models Desc": "",
"Choice Open": "Open",
"Choice Open Desc": "",
"Choice QA": "QA Struceture",
"Choice QA Desc": "The index is constructed with the structure of QA pairs, and ADAPTS to various scenarios such as Q&A and reading",
"Choice Visual": "Visual workflow",
"Choice Visual Desc": "Visualize modular operations, easily implement complex workflows, and make your AI no longer monolithic",
"Community": "Community",
"Dateset": "",
"Dateset Desc": "",
"Docs": "Docs",
"FastGPT Ability": "FastGPT Ability",
"FastGPT Desc": "FastGPT is a knowledgebase question answering system based on LLM large language model, which provides out-of-the-box data processing, model invocation and other capabilities. At the same time, workflow orchestration can be performed through Flow visualization to achieve complex Q&A scenarios!",
"Features": "Features",
"Footer Developer": "Developer",
"Footer Docs": "Docs",
"Footer FastGPT Cloud": "FastGPT Cloud",
"Footer Feedback": "Feedback",
"Footer Git": "Code",
"Footer Product": "Product",
"Footer Support": "Support",
"Login": "Login",
"Open": "",
"OpenAPI": "OpenAPI",
"OpenAPI Desc": "",
"Quickly build AI question and answer library": "Quickly build AI question and answer library",
"Start Now": "Start Now",
"Visual AI orchestration": "Visual AI orchestration",
"Why FastGPT": "",
"desc": "AI knowledge base question and answer platform based on LLM large model",
"slogan": "Let the AI know more about you"
},
"navbar": {
"Account": "Account",
"Apps": "Apps",
"Chat": "Chat",
"Datasets": "DataSets",
"Store": "Store",
"Tools": "Tools"
},
"user": {
"Account": "Account",
"Amount of earnings": "Earnings",
"Amount of inviter": "Inviter",
"Application Name": "Application Name",
"Avatar": "Avatar",
"Balance": "Balance",
"Bill Detail": "Bill Detail",
"Change": "Change",
"Copy invite url": "Copy invitation link",
"Invite Url": "Invite Url",
"Invite url tip": "Friends who register through this link will be permanently bound to you, and you will get a certain balance reward when they recharge. In addition, when friends register with their mobile phone number, you will get 5 yuan reward immediately.",
"Notice": "Notice",
"Old password is error": "Old password is error",
"OpenAI Account Setting": "OpenAI Account Setting",
"Password": "Password",
"Pay": "Pay",
"Personal Information": "Personal",
"Promotion": "Promotion",
"Promotion Rate": "Promotion Rate",
"Promotion Record": "Promotion",
"Promotion rate tip": "You will be rewarded with a percentage of the balance when your friends top up",
"Recharge Record": "Recharge",
"Replace": "Replace",
"Set OpenAI Account Failed": "Set OpenAI account failed",
"Sign Out": "Sign Out",
"Source": "Source",
"Time": "Time",
"Total Amount": "Total Amount",
"Update Password": "Update Password",
"Update password failed": "Update password failed",
"Update password succseful": "Update password succseful",
"Usage Record": "Usage",
"promotion": {
"pay": "",
"register": ""
}
}
}

View File

@@ -1,191 +0,0 @@
{
"App": "应用",
"Cancel": "取消",
"Confirm": "确认",
"Running": "运行中",
"Select value is empty": "选择的内容为空",
"UnKnow": "未知",
"Warning": "提示",
"app": {
"Advance App TestTip": "当前应用为高级编排模式\n如需切换为【简易模式】请点击左侧保存按键",
"App Detail": "应用详情",
"Chat Logs Tips": "日志会记录该应用的在线、分享和 API 对话记录",
"Chat logs": "对话日志",
"Confirm Del App Tip": "确认删除该应用及其所有聊天记录?",
"Confirm Save App Tip": "该应用可能为高级编排模式,保存后将会覆盖高级编排配置,请确认!",
"Connection is invalid": "连接无效",
"Connection type is different": "连接的类型不一致",
"Copy Module Config": "复制配置",
"Export Config Successful": "已复制配置,请注意检查是否有重要数据",
"Export Configs": "导出配置",
"Feedback Count": "用户反馈",
"Import Config": "导入配置",
"Import Config Failed": "导入配置失败,请确保配置正常!",
"Import Configs": "导入配置",
"Input Field Settings": "输入字段编辑",
"Logs Empty": "还没有日志噢~",
"Logs Message Total": "消息总数",
"Logs Source": "来源",
"Logs Time": "时间",
"Logs Title": "标题",
"Mark Count": "标注答案数量",
"My Apps": "我的应用",
"Output Field Settings": "输出字段编辑",
"Paste Config": "粘贴配置"
},
"chat": {
"Admin Mark Content": "纠正后的回复",
"Complete Response": "完整响应",
"Confirm to clear history": "确认清空该应用的聊天记录?",
"Exit Chat": "退出聊天",
"Feedback Failed": "提交反馈异常",
"Feedback Mark": "标注",
"Feedback Modal": "结果反馈",
"Feedback Modal Tip": "输入你觉得回答不满意的地方",
"Feedback Close": "关闭反馈",
"Feedback Submit": "提交反馈",
"Feedback Success": "反馈成功!",
"Feedback Update Failed": "更新反馈状态失败",
"History": "记录",
"Mark": "标注预期回答",
"Mark Description": "当前标注功能为测试版。\n\n点击添加标注后需要选择一个知识库以便存储标注数据。你可以通过该功能快速的标注问题和预期回答以便引导模型下次的回答。\n\n目前标注功能同知识库其他数据一样受模型的影响不代表标注后 100% 符合预期。\n\n标注数据仅单向与知识库同步如果知识库修改了该标注数据日志展示的标注数据无法同步",
"Mark Description Title": "标注功能介绍",
"New Chat": "新对话",
"Read Mark Description": "查看标注功能介绍",
"Read User Feedback": "查看用户反馈",
"Select Mark Kb": "选择知识库",
"Select Mark Kb Desc": "选择一个知识库存储预期答案",
"You need to a chat app": "你需要创建一个应用",
"logs": {
"api": "API 调用",
"online": "在线使用",
"share": "外部链接调用",
"test": "测试"
}
},
"commom": {
"Password inconsistency": "两次密码不一致"
},
"common": {
"Add": "添加",
"Cancel": "取消",
"Collect": "收藏",
"Copy": "复制",
"Copy Successful": "复制成功",
"Course": "",
"Delete": "删除",
"Filed is repeat": "",
"Filed is repeated": "字段重复了",
"Input": "输入",
"Output": "输出",
"export": ""
},
"dataset": {
"Confirm to delete the data": "确认删除该数据?",
"Queue Desc": "该数据是指整个系统当前待训练的数量。FastGPT 采用排队训练的方式,如果待训练的数据过多,可能需要等待一段时间",
"System Data Queue": "排队长度"
},
"file": {
"Click to download CSV template": "点击下载 CSV 模板",
"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",
"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",
"max 10": "最多选择 10 个文件",
"select a document": "选择文件",
"support": "支持 {{fileExtension}} 文件",
"upload error description": "单次只支持上传多个文件或者一个文件夹"
},
"home": {
"AI Assistant": "AI 客服",
"AI Assistant Desc": "无论对内还是对外AI 将 24 小时为您的用户提供服务",
"Advanced Settings": "高级编排",
"Advanced Settings Desc": "基于 Flow 的流程编排模式,让你的 AI 轻松实现数据库查询、IO 操作、联网通信等扩展能力",
"Choice Debug": "调试便捷",
"Choice Debug Desc": "拥有搜索测试、引用修改、完整对话预览等多种调试途径",
"Choice Desc": "",
"Choice Extension": "无限扩展",
"Choice Extension Desc": "基于 HTTP 实现扩展,轻松实现定制功能",
"Choice Models": "支持多种模型",
"Choice Models Desc": "支持 GPT、Claude、文心一言等多模型",
"Choice Open": "更开放",
"Choice Open Desc": "FastGPT 遵循 Apache License 2.0 开源协议",
"Choice QA": "独特的 QA 结构",
"Choice QA Desc": "采用 QA 对的结构构建索引,适应问答、阅读等多种场景",
"Choice Visual": "可视化工作流",
"Choice Visual Desc": "可视化模块操作,轻松实现复杂工作流,让你的 AI 不再单一",
"Community": "社区",
"Dateset": "自动数据预处理",
"Dateset Desc": "提供手动输入、直接分段、LLM 自动处理和 CSV 等多种数据导入途径",
"Docs": "文档",
"FastGPT Ability": "FastGPT 能力",
"FastGPT Desc": "FastGPT 是一个基于 LLM 大语言模型的知识库问答系统,提供开箱即用的数据处理、模型调用等能力。同时可以通过 Flow 可视化进行工作流编排,从而实现复杂的问答场景!",
"Features": "特点",
"Footer Developer": "开发者",
"Footer Docs": "文档",
"Footer FastGPT Cloud": "FastGPT 线上服务",
"Footer Feedback": "反馈",
"Footer Git": "源码",
"Footer Product": "产品",
"Footer Support": "支持",
"Login": "登录",
"Open": "",
"OpenAPI": "OpenAPI",
"OpenAPI Desc": "与 GPT API 一致的对外接口,助你轻松接入已有应用",
"Quickly build AI question and answer library": "快速搭建 AI 问答系统",
"Start Now": "立即开始",
"Visual AI orchestration": "可视化 AI 编排",
"Why FastGPT": "为什么选择 FastGPT",
"desc": "基于 LLM 大模型的 AI 知识库问答平台",
"slogan": "让 AI 更懂你的知识"
},
"navbar": {
"Account": "账号",
"Apps": "应用",
"Chat": "聊天",
"Datasets": "知识库",
"Store": "应用市场",
"Tools": "工具"
},
"user": {
"Account": "账号",
"Amount of earnings": "收益(¥)",
"Amount of inviter": "累计邀请人数",
"Application Name": "应用名",
"Avatar": "头像",
"Balance": "余额",
"Bill Detail": "账单详情",
"Change": "变更",
"Copy invite url": "复制邀请链接",
"Invite Url": "邀请链接",
"Invite url tip": "通过该链接注册的好友将永久与你绑定,其充值时你会获得一定余额奖励。\n此外好友使用手机号注册时你将立即获得 5 元奖励。",
"Notice": "通知",
"Old password is error": "旧密码错误",
"OpenAI Account Setting": "OpenAI 账号配置",
"Password": "密码",
"Pay": "充值",
"Personal Information": "个人信息",
"Promotion": "",
"Promotion Rate": "返现比例",
"Promotion Record": "推广记录",
"Promotion rate tip": "好友充值时你将获得一定比例的余额奖励",
"Recharge Record": "充值记录",
"Replace": "更换",
"Set OpenAI Account Failed": "设置 OpenAI 账号异常",
"Sign Out": "登出",
"Source": "来源",
"Time": "时间",
"Total Amount": "总金额",
"Update Password": "修改密码",
"Update password failed": "修改密码异常",
"Update password succseful": "修改密码成功",
"Usage Record": "使用记录",
"promotion": {
"pay": "好友充值",
"register": "好友注册"
}
}
}

View File

@@ -1,16 +0,0 @@
import { GET, POST, DELETE } from './request';
import { UserOpenApiKey } from '@/types/openapi';
/**
* crete a api key
*/
export const createAOpenApiKey = () => POST<string>('/openapi/postKey');
/**
* get api keys
*/
export const getOpenApiKeys = () => GET<UserOpenApiKey[]>('/openapi/getKeys');
/**
* delete api by id
*/
export const delOpenApiById = (id: string) => DELETE(`/openapi/delKey?id=${id}`);

View File

@@ -1,6 +0,0 @@
import { GET, POST, PUT, DELETE } from '../request';
import type { FetchResultItem } from '@/types/plugin';
export const fetchUrls = (urlList: string[]) =>
POST<FetchResultItem[]>(`/plugins/urlFetch`, { urlList });

View File

@@ -1,98 +0,0 @@
import { GET, POST, PUT, DELETE } from '../request';
import type { KbItemType, KbListItemType } from '@/types/plugin';
import { RequestPaging } from '@/types/index';
import { TrainingModeEnum } from '@/constants/plugin';
import {
Props as PushDataProps,
Response as PushDateResponse
} from '@/pages/api/openapi/kb/pushData';
import {
Props as SearchTestProps,
Response as SearchTestResponse
} from '@/pages/api/openapi/kb/searchTest';
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';
/* knowledge base */
export const getKbList = () => GET<KbListItemType[]>(`/plugins/kb/list`);
export const getKbById = (id: string) => GET<KbItemType>(`/plugins/kb/detail?id=${id}`);
export const postCreateKb = (data: CreateKbParams) => POST<string>(`/plugins/kb/create`, data);
export const putKbById = (data: KbUpdateParams) => PUT(`/plugins/kb/update`, data);
export const delKbById = (id: string) => DELETE(`/plugins/kb/delete?id=${id}`);
/* kb data */
type GetKbDataListProps = RequestPaging & {
kbId: string;
searchText: string;
};
export const getKbDataList = (data: GetKbDataListProps) =>
POST(`/plugins/kb/data/getDataList`, data);
/**
* 获取导出数据(不分页)
*/
export const getExportDataList = (kbId: string) =>
GET<[string, string, string][]>(
`/plugins/kb/data/exportModelData`,
{ kbId },
{
timeout: 600000
}
);
/**
* 获取模型正在拆分数据的数量
*/
export const getTrainingData = (data: { kbId: string; init: boolean }) =>
POST<{
qaListLen: number;
vectorListLen: number;
}>(`/plugins/kb/data/getTrainingData`, data);
/* get length of system training queue */
export const getTrainingQueueLen = () => GET<number>(`/plugins/kb/data/getQueueLen`);
export const getKbDataItemById = (dataId: string) =>
GET<KbDataItemType>(`/plugins/kb/data/getDataById`, { dataId });
/**
* 直接push数据
*/
export const postKbDataFromList = (data: PushDataProps) =>
POST<PushDateResponse>(`/openapi/kb/pushData`, data);
/**
* 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);
/**
* 更新一条数据
*/
export const putKbDataById = (data: UpdateDataProps) => PUT('/openapi/kb/updateData', data);
/**
* 删除一条知识库数据
*/
export const delOneKbDataByDataId = (dataId: string) =>
DELETE(`/openapi/kb/delDataById?dataId=${dataId}`);
/**
* 拆分数据
*/
export const postSplitData = (data: {
kbId: string;
chunks: string[];
prompt: string;
mode: `${TrainingModeEnum}`;
}) => POST(`/openapi/text/pushData`, data);
export const searchText = (data: SearchTestProps) =>
POST<SearchTestResponse>(`/openapi/kb/searchTest`, data);

View File

@@ -1,6 +0,0 @@
export type AdminUpdateFeedbackParams = {
chatItemId: string;
kbId: string;
dataId: string;
content: string;
};

View File

@@ -1,12 +0,0 @@
export type KbUpdateParams = {
id: string;
name: string;
tags: string;
avatar: string;
};
export type CreateKbParams = {
name: string;
tags: string[];
avatar: string;
vectorModel: string;
};

View File

@@ -1,8 +0,0 @@
import { GET, POST } from './request';
export const textCensor = (data: { text: string }) =>
POST<{ code?: number; message: string }>('/plugins/censor/text_baidu', data).then((res) => {
if (res?.code === 5000) {
return Promise.reject(res.message);
}
});

View File

@@ -1,6 +0,0 @@
import { GET, POST, PUT } from './request';
import type { InitDateResponse } from '@/pages/api/system/getInitData';
export const getInitData = () => GET<InitDateResponse>('/system/getInitData');
export const uploadImg = (base64Img: string) => POST<string>('/system/uploadImage', { base64Img });

View File

@@ -1,143 +0,0 @@
import React, { useState } from 'react';
import {
Box,
Button,
Flex,
ModalFooter,
ModalBody,
Table,
Thead,
Tbody,
Tr,
Th,
Td,
TableContainer,
IconButton
} from '@chakra-ui/react';
import { getOpenApiKeys, createAOpenApiKey, delOpenApiById } from '@/api/openapi';
import { useQuery, useMutation } from '@tanstack/react-query';
import { useLoading } from '@/hooks/useLoading';
import dayjs from 'dayjs';
import { AddIcon, DeleteIcon } from '@chakra-ui/icons';
import { getErrText, useCopyData } from '@/utils/tools';
import { useToast } from '@/hooks/useToast';
import MyIcon from '../Icon';
import MyModal from '../MyModal';
const APIKeyModal = ({ onClose }: { onClose: () => void }) => {
const { Loading } = useLoading();
const { toast } = useToast();
const {
data: apiKeys = [],
isLoading: isGetting,
refetch
} = useQuery(['getOpenApiKeys'], getOpenApiKeys);
const [apiKey, setApiKey] = useState('');
const { copyData } = useCopyData();
const { mutate: onclickCreateApiKey, isLoading: isCreating } = useMutation({
mutationFn: () => createAOpenApiKey(),
onSuccess(res) {
setApiKey(res);
refetch();
},
onError(err) {
toast({
status: 'warning',
title: getErrText(err)
});
}
});
const { mutate: onclickRemove, isLoading: isDeleting } = useMutation({
mutationFn: async (id: string) => delOpenApiById(id),
onSuccess() {
refetch();
}
});
return (
<MyModal isOpen onClose={onClose} w={'600px'}>
<Box py={3} px={5}>
<Box fontWeight={'bold'} fontSize={'2xl'}>
API
</Box>
<Box fontSize={'sm'} color={'myGray.600'}>
API 使~
</Box>
</Box>
<ModalBody minH={'300px'} maxH={['70vh', '500px']} overflow={'overlay'}>
<TableContainer mt={2} position={'relative'}>
<Table>
<Thead>
<Tr>
<Th>Api Key</Th>
<Th></Th>
<Th>使</Th>
<Th />
</Tr>
</Thead>
<Tbody fontSize={'sm'}>
{apiKeys.map(({ id, apiKey, createTime, lastUsedTime }) => (
<Tr key={id}>
<Td>{apiKey}</Td>
<Td>{dayjs(createTime).format('YYYY/MM/DD HH:mm:ss')}</Td>
<Td>
{lastUsedTime
? dayjs(lastUsedTime).format('YYYY/MM/DD HH:mm:ss')
: '没有使用过'}
</Td>
<Td>
<IconButton
icon={<DeleteIcon />}
size={'xs'}
aria-label={'delete'}
variant={'base'}
colorScheme={'gray'}
onClick={() => onclickRemove(id)}
/>
</Td>
</Tr>
))}
</Tbody>
</Table>
</TableContainer>
</ModalBody>
<ModalFooter>
<Button
variant="base"
leftIcon={<AddIcon color={'myGray.600'} fontSize={'sm'} />}
onClick={() => onclickCreateApiKey()}
>
</Button>
</ModalFooter>
<Loading loading={isGetting || isCreating || isDeleting} fixed={false} />
<MyModal isOpen={!!apiKey} w={'400px'} onClose={() => setApiKey('')}>
<Box py={3} px={5}>
<Box fontWeight={'bold'} fontSize={'2xl'}>
API
</Box>
<Box fontSize={'sm'} color={'myGray.600'}>
~
</Box>
</Box>
<ModalBody>
<Flex bg={'myGray.100'} px={3} py={2} cursor={'pointer'} onClick={() => copyData(apiKey)}>
<Box flex={1}>{apiKey}</Box>
<MyIcon name={'copy'} w={'16px'}></MyIcon>
</Flex>
</ModalBody>
<ModalFooter>
<Button variant="base" onClick={() => setApiKey('')}>
</Button>
</ModalFooter>
</MyModal>
</MyModal>
);
};
export default APIKeyModal;

View File

@@ -1,149 +0,0 @@
import React, { useCallback, useState } from 'react';
import { ModalBody, Box, useTheme } from '@chakra-ui/react';
import { getKbDataItemById } from '@/api/plugins/kb';
import { useLoading } from '@/hooks/useLoading';
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 MyModal from '../MyModal';
type SearchType = {
kb_id?: string;
id?: string;
q: string;
a?: string;
source?: string | undefined;
};
const QuoteModal = ({
onUpdateQuote,
rawSearch = [],
onClose
}: {
onUpdateQuote: (quoteId: string, sourceText: string) => Promise<void>;
rawSearch: SearchType[];
onClose: () => void;
}) => {
const theme = useTheme();
const { toast } = useToast();
const { setIsLoading, Loading } = useLoading();
const [editDataItem, setEditDataItem] = useState<{
kbId: string;
dataId: string;
a: string;
q: string;
}>();
/**
* click edit, get new kbDataItem
*/
const onclickEdit = useCallback(
async (item: SearchType) => {
if (!item.id) return;
try {
setIsLoading(true);
const data = (await getKbDataItemById(item.id)) as QuoteItemType;
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
});
} catch (err) {
toast({
status: 'warning',
title: getErrText(err)
});
}
setIsLoading(false);
},
[setIsLoading, toast, onUpdateQuote]
);
return (
<>
<MyModal
isOpen={true}
onClose={onClose}
h={['90vh', '80vh']}
isCentered
minW={['90vw', '600px']}
title={
<>
({rawSearch.length})
<Box fontSize={['xs', 'sm']} fontWeight={'normal'}>
注意: 修改知识库内容成功后
</Box>
</>
}
>
<ModalBody pt={0} whiteSpace={'pre-wrap'} textAlign={'justify'} fontSize={'sm'}>
{rawSearch.map((item, i) => (
<Box
key={i}
flex={'1 0 0'}
p={2}
borderRadius={'lg'}
border={theme.borders.base}
_notLast={{ mb: 2 }}
position={'relative'}
_hover={{ '& .edit': { display: 'flex' } }}
overflow={'hidden'}
>
{item.source && <Box color={'myGray.600'}>({item.source})</Box>}
<Box>{item.q}</Box>
<Box>{item.a}</Box>
{item.id && (
<Box
className="edit"
display={'none'}
position={'absolute'}
right={0}
top={0}
bottom={0}
w={'40px'}
bg={'rgba(255,255,255,0.9)'}
alignItems={'center'}
justifyContent={'center'}
boxShadow={'-10px 0 10px rgba(255,255,255,1)'}
>
<MyIcon
name={'edit'}
w={'18px'}
h={'18px'}
cursor={'pointer'}
color={'myGray.600'}
_hover={{
color: 'myBlue.700'
}}
onClick={() => onclickEdit(item)}
/>
</Box>
)}
</Box>
))}
</ModalBody>
<Loading fixed={false} />
</MyModal>
{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}
/>
)}
</>
);
};
export default QuoteModal;

View File

@@ -1,108 +0,0 @@
import React, { useCallback, useMemo, useState } from 'react';
import { ChatModuleEnum } from '@/constants/chat';
import { ChatHistoryItemResType, ChatItemType, QuoteItemType } from '@/types/chat';
import { Flex, BoxProps, useDisclosure } from '@chakra-ui/react';
import { useTranslation } from 'react-i18next';
import { useGlobalStore } from '@/store/global';
import dynamic from 'next/dynamic';
import Tag from '../Tag';
import MyTooltip from '../MyTooltip';
const QuoteModal = dynamic(() => import('./QuoteModal'), { ssr: false });
const ContextModal = dynamic(() => import('./ContextModal'), { ssr: false });
const WholeResponseModal = dynamic(() => import('./WholeResponseModal'), { ssr: false });
const ResponseTags = ({
chatId,
contentId,
responseData = []
}: {
chatId?: string;
contentId?: string;
responseData?: ChatHistoryItemResType[];
}) => {
const { isPc } = useGlobalStore();
const { t } = useTranslation();
const [quoteModalData, setQuoteModalData] = useState<QuoteItemType[]>();
const [contextModalData, setContextModalData] = useState<ChatItemType[]>();
const {
isOpen: isOpenWholeModal,
onOpen: onOpenWholeModal,
onClose: onCloseWholeModal
} = useDisclosure();
const {
quoteList = [],
completeMessages = [],
tokens = 0
} = useMemo(() => {
const chatData = responseData.find((item) => item.moduleName === ChatModuleEnum.AIChat);
if (!chatData) return {};
return {
quoteList: chatData.quoteList,
completeMessages: chatData.completeMessages,
tokens: responseData.reduce((sum, item) => sum + (item.tokens || 0), 0)
};
}, [responseData]);
const updateQuote = useCallback(async (quoteId: string, sourceText: string) => {}, []);
const TagStyles: BoxProps = {
mr: 2,
bg: 'transparent'
};
return responseData.length === 0 ? null : (
<Flex alignItems={'center'} mt={2} flexWrap={'wrap'}>
{quoteList.length > 0 && (
<MyTooltip label="查看引用">
<Tag
colorSchema="blue"
cursor={'pointer'}
{...TagStyles}
onClick={() => setQuoteModalData(quoteList)}
>
{quoteList.length}
</Tag>
</MyTooltip>
)}
{completeMessages.length > 0 && (
<MyTooltip label={'点击查看完整对话记录'}>
<Tag
colorSchema="green"
cursor={'pointer'}
{...TagStyles}
onClick={() => setContextModalData(completeMessages)}
>
{completeMessages.length}
</Tag>
</MyTooltip>
)}
{isPc && tokens > 0 && (
<Tag colorSchema="purple" cursor={'default'} {...TagStyles}>
{tokens}Tokens
</Tag>
)}
<MyTooltip label={'点击查看完整响应值'}>
<Tag colorSchema="gray" cursor={'pointer'} {...TagStyles} onClick={onOpenWholeModal}>
{t('chat.Complete Response')}
</Tag>
</MyTooltip>
{!!quoteModalData && (
<QuoteModal
rawSearch={quoteModalData}
onUpdateQuote={updateQuote}
onClose={() => setQuoteModalData(undefined)}
/>
)}
{!!contextModalData && (
<ContextModal context={contextModalData} onClose={() => setContextModalData(undefined)} />
)}
{isOpenWholeModal && (
<WholeResponseModal response={responseData} onClose={onCloseWholeModal} />
)}
</Flex>
);
};
export default ResponseTags;

View File

@@ -1,117 +0,0 @@
import React, { useRef, useState } from 'react';
import {
ModalBody,
useTheme,
ModalFooter,
Button,
ModalHeader,
Box,
Card,
Flex
} from '@chakra-ui/react';
import MyModal from '../MyModal';
import { useTranslation } from 'next-i18next';
import { useQuery } from '@tanstack/react-query';
import { useUserStore } from '@/store/user';
import { useToast } from '@/hooks/useToast';
import Avatar from '../Avatar';
import MyIcon from '@/components/Icon';
import { useGlobalStore } from '@/store/global';
const SelectDataset = ({
onSuccess,
onClose
}: {
onSuccess: (kbId: string) => void;
onClose: () => void;
}) => {
const { t } = useTranslation();
const theme = useTheme();
const { isPc } = useGlobalStore();
const { toast } = useToast();
const { myKbList, loadKbList } = useUserStore();
const [selectedId, setSelectedId] = useState<string>();
useQuery(['loadKbList'], loadKbList);
return (
<MyModal isOpen={true} onClose={onClose} w={'100%'} maxW={['90vw', '900px']} isCentered={!isPc}>
<Flex flexDirection={'column'} h={['90vh', 'auto']}>
<ModalHeader>
<Box>{t('chat.Select Mark Kb')}</Box>
<Box fontSize={'sm'} color={'myGray.500'} fontWeight={'normal'}>
{t('chat.Select Mark Kb Desc')}
</Box>
</ModalHeader>
<ModalBody
flex={['1 0 0', '0 0 auto']}
maxH={'80vh'}
overflowY={'auto'}
display={'grid'}
gridTemplateColumns={['repeat(1,1fr)', 'repeat(2,1fr)', 'repeat(3,1fr)']}
gridGap={3}
userSelect={'none'}
>
{myKbList.map((item) =>
(() => {
const selected = selectedId === item._id;
return (
<Card
key={item._id}
p={3}
border={theme.borders.base}
boxShadow={'sm'}
h={'80px'}
cursor={'pointer'}
_hover={{
boxShadow: 'md'
}}
{...(selected
? {
bg: 'myBlue.300'
}
: {})}
onClick={() => {
setSelectedId(item._id);
}}
>
<Flex alignItems={'center'} h={'38px'}>
<Avatar src={item.avatar} w={['24px', '28px', '32px']}></Avatar>
<Box ml={3} fontWeight={'bold'} fontSize={['md', 'lg', 'xl']}>
{item.name}
</Box>
</Flex>
<Flex justifyContent={'flex-end'} alignItems={'center'} fontSize={'sm'}>
<MyIcon mr={1} name="kbTest" w={'12px'} />
<Box color={'myGray.500'}>{item.vectorModel.name}</Box>
</Flex>
</Card>
);
})()
)}
</ModalBody>
<ModalFooter>
<Button variant={'base'} mr={2} onClick={onClose}>
{t('Cancel')}
</Button>
<Button
onClick={() => {
if (!selectedId) {
return toast({
status: 'warning',
title: t('Select value is empty')
});
}
onSuccess(selectedId);
}}
>
{t('Confirm')}
</Button>
</ModalFooter>
</Flex>
</MyModal>
);
};
export default SelectDataset;

View File

@@ -1,71 +0,0 @@
import React, { useMemo } from 'react';
import { Box, ModalBody, useTheme, ModalHeader, Flex } from '@chakra-ui/react';
import type { ChatHistoryItemResType } from '@/types/chat';
import { useTranslation } from 'react-i18next';
import MyModal from '../MyModal';
import MyTooltip from '../MyTooltip';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
const ResponseModal = ({
response,
onClose
}: {
response: ChatHistoryItemResType[];
onClose: () => void;
}) => {
const { t } = useTranslation();
const theme = useTheme();
const formatResponse = useMemo(
() =>
response.map((item) => {
const copy = { ...item };
delete copy.completeMessages;
delete copy.quoteList;
return copy;
}),
[response]
);
return (
<MyModal
isOpen={true}
onClose={onClose}
h={['90vh', '80vh']}
minW={['90vw', '600px']}
title={
<Flex alignItems={'center'}>
{t('chat.Complete Response')}
<MyTooltip
label={
'moduleName: 模型名\nprice: 价格倍率100000\nmodel?: 模型名\ntokens?: token 消耗\n\nanswer?: 回答内容\nquestion?: 问题\ntemperature?: 温度\nmaxToken?: 最大 tokens\n\nsimilarity?: 相似度\nlimit?: 单次搜索结果\n\ncqList?: 问题分类列表\ncqResult?: 分类结果\n\nextractDescription?: 内容提取描述\nextractResult?: 提取结果'
}
>
<QuestionOutlineIcon ml={2} />
</MyTooltip>
</Flex>
}
isCentered
>
<ModalBody>
{formatResponse.map((item, i) => (
<Box
key={i}
p={2}
pt={[0, 2]}
borderRadius={'lg'}
border={theme.borders.base}
_notLast={{ mb: 2 }}
position={'relative'}
whiteSpace={'pre-wrap'}
>
{JSON.stringify(item, null, 2)}
</Box>
))}
</ModalBody>
</MyModal>
);
};
export default ResponseModal;

View File

@@ -1,29 +0,0 @@
import { SystemInputEnum } from '@/constants/app';
import { FlowModuleTypeEnum } from '@/constants/flow';
import { getChatModel } from '@/service/utils/data';
import { AppModuleItemType, VariableItemType } from '@/types/app';
export const getSpecialModule = (modules: AppModuleItemType[]) => {
const welcomeText: string =
modules
.find((item) => item.flowType === FlowModuleTypeEnum.userGuide)
?.inputs?.find((item) => item.key === SystemInputEnum.welcomeText)?.value || '';
const variableModules: VariableItemType[] =
modules
.find((item) => item.flowType === FlowModuleTypeEnum.variable)
?.inputs.find((item) => item.key === SystemInputEnum.variables)?.value || [];
return {
welcomeText,
variableModules
};
};
export const getChatModelNameList = (modules: AppModuleItemType[]): string[] => {
const chatModules = modules.filter((item) => item.flowType === FlowModuleTypeEnum.chatNode);
return chatModules
.map(
(item) => getChatModel(item.inputs.find((input) => input.key === 'model')?.value)?.name || ''
)
.filter((item) => item);
};

View File

@@ -1,4 +0,0 @@
.datePicker {
--rdp-background-color: #d6e8ff;
--rdp-accent-color: #0000ff;
}

View File

@@ -1,104 +0,0 @@
import React from 'react';
import type { IconProps } from '@chakra-ui/react';
import { Icon } from '@chakra-ui/react';
const map = {
appFill: require('./icons/fill/app.svg').default,
appLight: require('./icons/light/app.svg').default,
copy: require('./icons/copy.svg').default,
chatSend: require('./icons/chatSend.svg').default,
delete: require('./icons/delete.svg').default,
stop: require('./icons/stop.svg').default,
collectionLight: require('./icons/collectionLight.svg').default,
collectionSolid: require('./icons/collectionSolid.svg').default,
empty: require('./icons/empty.svg').default,
back: require('./icons/back.svg').default,
backFill: require('./icons/fill/back.svg').default,
more: require('./icons/more.svg').default,
tabbarChat: require('./icons/phoneTabbar/chat.svg').default,
tabbarModel: require('./icons/phoneTabbar/app.svg').default,
tabbarMore: require('./icons/phoneTabbar/more.svg').default,
tabbarMe: require('./icons/phoneTabbar/me.svg').default,
closeSolid: require('./icons/closeSolid.svg').default,
wx: require('./icons/wx.svg').default,
out: require('./icons/out.svg').default,
git: require('./icons/git.svg').default,
gitFill: require('./icons/fill/git.svg').default,
menu: require('./icons/menu.svg').default,
edit: require('./icons/edit.svg').default,
inform: require('./icons/inform.svg').default,
export: require('./icons/export.svg').default,
text: require('./icons/text.svg').default,
history: require('./icons/history.svg').default,
kbTest: require('./icons/kbTest.svg').default,
date: require('./icons/date.svg').default,
apikey: require('./icons/apikey.svg').default,
save: require('./icons/save.svg').default,
minus: require('./icons/minus.svg').default,
chat: require('./icons/light/chat.svg').default,
chatFill: require('./icons/fill/chat.svg').default,
clear: require('./icons/light/clear.svg').default,
apiLight: require('./icons/light/appApi.svg').default,
overviewLight: require('./icons/light/overview.svg').default,
settingLight: require('./icons/light/setting.svg').default,
shareLight: require('./icons/light/share.svg').default,
dbLight: require('./icons/light/db.svg').default,
dbFill: require('./icons/fill/db.svg').default,
appStoreLight: require('./icons/light/appStore.svg').default,
appStoreFill: require('./icons/fill/appStore.svg').default,
meLight: require('./icons/light/me.svg').default,
meFill: require('./icons/fill/me.svg').default,
welcomeText: require('./icons/modules/welcomeText.svg').default,
variable: require('./icons/modules/variable.svg').default,
setTop: require('./icons/light/setTop.svg').default,
fullScreenLight: require('./icons/light/fullScreen.svg').default,
voice: require('./icons/voice.svg').default,
html: require('./icons/file/html.svg').default,
pdf: require('./icons/file/pdf.svg').default,
markdown: require('./icons/file/markdown.svg').default,
importLight: require('./icons/light/import.svg').default,
manualImport: require('./icons/file/manualImport.svg').default,
indexImport: require('./icons/file/indexImport.svg').default,
csvImport: require('./icons/file/csv.svg').default,
qaImport: require('./icons/file/qaImport.svg').default,
uploadFile: require('./icons/file/uploadFile.svg').default,
closeLight: require('./icons/light/close.svg').default,
customTitle: require('./icons/light/customTitle.svg').default,
billRecordLight: require('./icons/light/billRecord.svg').default,
informLight: require('./icons/light/inform.svg').default,
payRecordLight: require('./icons/light/payRecord.svg').default,
loginoutLight: require('./icons/light/loginout.svg').default,
chatModelTag: require('./icons/light/chatModelTag.svg').default,
language_en: require('./icons/language/en.svg').default,
language_zh: require('./icons/language/zh.svg').default,
outlink_share: require('./icons/outlink/share.svg').default,
outlink_iframe: require('./icons/outlink/iframe.svg').default,
addCircle: require('./icons/circle/add.svg').default,
playFill: require('./icons/fill/play.svg').default,
courseLight: require('./icons/light/course.svg').default,
promotionLight: require('./icons/light/promotion.svg').default,
logsLight: require('./icons/light/logs.svg').default,
badLight: require('./icons/light/bad.svg').default,
markLight: require('./icons/light/mark.svg').default
};
export type IconName = keyof typeof map;
const MyIcon = (
{ name, w = 'auto', h = 'auto', ...props }: { name: IconName } & IconProps,
ref: any
) => {
return map[name] ? (
<Icon
as={map[name]}
w={w}
h={h}
boxSizing={'content-box'}
verticalAlign={'top'}
fill={'currentcolor'}
{...props}
/>
) : null;
};
export default React.forwardRef(MyIcon);

View File

@@ -1,22 +0,0 @@
import React from 'react';
import { Spinner, Flex } from '@chakra-ui/react';
const Loading = ({ fixed = true }: { fixed?: boolean }) => {
return (
<Flex
position={fixed ? 'fixed' : 'absolute'}
zIndex={1000}
backgroundColor={'rgba(255,255,255,0.5)'}
top={0}
left={0}
right={0}
bottom={0}
alignItems={'center'}
justifyContent={'center'}
>
<Spinner thickness="4px" speed="0.65s" emptyColor="myGray.100" color="myBlue.600" size="xl" />
</Flex>
);
};
export default Loading;

View File

@@ -1,49 +0,0 @@
import React, { useMemo } from 'react';
import { Box } from '@chakra-ui/react';
import ReactMarkdown from 'react-markdown';
import RemarkGfm from 'remark-gfm';
import RemarkMath from 'remark-math';
import RehypeKatex from 'rehype-katex';
import 'katex/dist/katex.min.css';
import styles from '../index.module.scss';
import { EventNameEnum } from '../constant';
const Guide = ({ text, onClick }: { text: string; onClick?: (e: any) => void }) => {
const formatText = useMemo(() => text.replace(/\[(.*?)\]/g, '[$1]()'), [text]);
return (
<ReactMarkdown
className={`markdown ${styles.markdown}`}
remarkPlugins={[RemarkGfm, RemarkMath]}
rehypePlugins={[RehypeKatex]}
components={{
a({ children }: any) {
return (
<Box as={'li'} py={1} m={0}>
<Box
as={'span'}
color={'blue.600'}
textDecoration={'underline'}
cursor={'pointer'}
onClick={() => {
if (!onClick) return;
onClick({
event: EventNameEnum.guideClick,
data: String(children)
});
}}
>
{String(children)}
</Box>
</Box>
);
}
}}
>
{formatText}
</ReactMarkdown>
);
};
export default React.memo(Guide);

View File

@@ -1,3 +0,0 @@
export enum EventNameEnum {
guideClick = 'guideClick'
}

View File

@@ -1,84 +0,0 @@
import type { BoxProps } from '@chakra-ui/react';
export enum FlowInputItemTypeEnum {
systemInput = 'systemInput', // history, userChatInput, variableInput
input = 'input',
textarea = 'textarea',
numberInput = 'numberInput',
select = 'select',
slider = 'slider',
custom = 'custom',
target = 'target',
none = 'none',
hidden = 'hidden'
}
export enum FlowOutputItemTypeEnum {
answer = 'answer',
source = 'source',
none = 'none',
hidden = 'hidden'
}
export enum FlowModuleTypeEnum {
empty = 'empty',
variable = 'variable',
userGuide = 'userGuide',
questionInput = 'questionInput',
historyNode = 'historyNode',
chatNode = 'chatNode',
kbSearchNode = 'kbSearchNode',
tfSwitchNode = 'tfSwitchNode',
answerNode = 'answerNode',
classifyQuestion = 'classifyQuestion',
contentExtract = 'contentExtract',
httpRequest = 'httpRequest'
}
export enum SpecialInputKeyEnum {
'answerText' = 'text',
'agents' = 'agents' // cq agent key
}
export enum FlowValueTypeEnum {
'string' = 'string',
'number' = 'number',
'boolean' = 'boolean',
'chatHistory' = 'chat_history',
'kbQuote' = 'kb_quote',
'any' = 'any'
}
export const FlowValueTypeStyle: Record<`${FlowValueTypeEnum}`, BoxProps> = {
[FlowValueTypeEnum.string]: {
background: '#36ADEF'
},
[FlowValueTypeEnum.number]: {
background: '#FB7C3C'
},
[FlowValueTypeEnum.boolean]: {
background: '#E7D118'
},
[FlowValueTypeEnum.chatHistory]: {
background: '#00A9A6'
},
[FlowValueTypeEnum.kbQuote]: {
background: '#A558C9'
},
[FlowValueTypeEnum.any]: {
background: '#9CA2A8'
}
};
export const initModuleType: Record<string, boolean> = {
[FlowModuleTypeEnum.historyNode]: true,
[FlowModuleTypeEnum.questionInput]: true
};
export const edgeOptions = {
style: {
strokeWidth: 1.5,
stroke: '#5A646Es'
}
};
export const connectionLineStyle = { strokeWidth: 1.5, stroke: '#5A646Es' };

View File

@@ -1,25 +0,0 @@
import { FlowInputItemType } from '@/types/flow';
import { SystemInputEnum } from '../app';
import { FlowInputItemTypeEnum, FlowValueTypeEnum } from './index';
export const Input_Template_TFSwitch: FlowInputItemType = {
key: SystemInputEnum.switch,
type: FlowInputItemTypeEnum.target,
label: '触发器',
valueType: FlowValueTypeEnum.any
};
export const Input_Template_History: FlowInputItemType = {
key: SystemInputEnum.history,
type: FlowInputItemTypeEnum.target,
label: '聊天记录',
valueType: FlowValueTypeEnum.chatHistory
};
export const Input_Template_UserChatInput: FlowInputItemType = {
key: SystemInputEnum.userChatInput,
type: FlowInputItemTypeEnum.target,
label: '用户问题',
required: true,
valueType: FlowValueTypeEnum.string
};

View File

@@ -1,29 +0,0 @@
import type { ShareChatEditType } from '@/types/app';
import type { AppSchema } from '@/types/mongoSchema';
export enum OpenAiChatEnum {
'GPT35' = 'gpt-3.5-turbo',
'GPT3516k' = 'gpt-3.5-turbo-16k',
'FastAI-Plus' = 'gpt-4',
'FastAI-Plus32k' = 'gpt-4-32k'
}
export const defaultApp: AppSchema = {
_id: '',
userId: 'userId',
name: '模型加载中',
type: 'basic',
avatar: '/icon/logo.svg',
intro: '',
updateTime: Date.now(),
share: {
isShare: false,
isShareDetail: false,
collection: 0
},
modules: []
};
export const defaultShareChat: ShareChatEditType = {
name: ''
};

View File

@@ -1,10 +0,0 @@
export enum TrainingModeEnum {
'qa' = 'qa',
'index' = 'index'
}
export const TrainingTypeMap = {
[TrainingModeEnum.qa]: 'qa',
[TrainingModeEnum.index]: 'index'
};
export const PgTrainingTableName = 'modeldata';

View File

@@ -1,19 +0,0 @@
import { useEffect } from 'react';
import { useRouter } from 'next/router';
function Error() {
const router = useRouter();
useEffect(() => {
setTimeout(() => {
router.replace('/app/list');
}, 2000);
}, []);
return (
<p>
safari chrome
</p>
);
}
export default Error;

View File

@@ -1,58 +0,0 @@
// 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 { connectToDatabase, Chat } from '@/service/mongo';
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
try {
await authUser({ req, authRoot: true });
await connectToDatabase();
const { limit = 1000 } = req.body as { limit: number };
let skip = 0;
const total = await Chat.countDocuments({
chatId: { $exists: false }
});
let promise = Promise.resolve();
console.log(total);
for (let i = 0; i < total; i += limit) {
const skipVal = skip;
skip += limit;
promise = promise
.then(() => init(limit, skipVal))
.then(() => {
console.log(skipVal);
});
}
await promise;
jsonRes(res, {});
} catch (error) {
jsonRes(res, {
code: 500,
error
});
}
}
async function init(limit: number, skip: number) {
// 遍历 app
const chats = await Chat.find(
{
chatId: { $exists: false }
},
'_id'
).limit(limit);
await Promise.all(
chats.map((chat) =>
Chat.findByIdAndUpdate(chat._id, {
chatId: String(chat._id),
source: 'online'
})
)
);
}

View File

@@ -1,98 +0,0 @@
// 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 { connectToDatabase, Chat, ChatItem } from '@/service/mongo';
import { customAlphabet } from 'nanoid';
const nanoid = customAlphabet('abcdefghijklmnopqrstuvwxyz1234567890', 24);
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
try {
await authUser({ req, authRoot: true });
await connectToDatabase();
const { limit = 100 } = req.body as { limit: number };
let skip = 0;
const total = await Chat.countDocuments({
content: { $exists: true, $not: { $size: 0 } },
isInit: { $ne: true }
});
const totalChat = await Chat.aggregate([
{
$project: {
contentLength: { $size: '$content' }
}
},
{
$group: {
_id: null,
totalLength: { $sum: '$contentLength' }
}
}
]);
console.log('chatLen:', total, totalChat);
let promise = Promise.resolve();
for (let i = 0; i < total; i += limit) {
const skipVal = skip;
skip += limit;
promise = promise
.then(() => init(limit))
.then(() => {
console.log(skipVal);
});
}
await promise;
jsonRes(res, {});
} catch (error) {
jsonRes(res, {
code: 500,
error
});
}
}
async function init(limit: number) {
// 遍历 app
const chats = await Chat.find(
{
content: { $exists: true, $not: { $size: 0 } },
isInit: { $ne: true }
},
'_id userId appId chatId content'
)
.sort({ updateTime: -1 })
.limit(limit);
await Promise.all(
chats.map(async (chat) => {
const inserts = chat.content
.map((item) => ({
dataId: nanoid(),
chatId: chat.chatId,
userId: chat.userId,
appId: chat.appId,
obj: item.obj,
value: item.value,
responseData: item.responseData
}))
.filter((item) => item.chatId && item.userId && item.appId && item.obj && item.value);
try {
await Promise.all(inserts.map((item) => ChatItem.create(item)));
await Chat.findByIdAndUpdate(chat._id, {
isInit: true
});
} catch (error) {
console.log(error);
await ChatItem.deleteMany({ chatId: chat.chatId });
}
})
);
}

View File

@@ -1,27 +0,0 @@
// 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 { connectToDatabase, OutLink } from '@/service/mongo';
import { OutLinkTypeEnum } from '@/constants/chat';
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
try {
await authUser({ req, authRoot: true });
await connectToDatabase();
await OutLink.updateMany(
{},
{
$set: { type: OutLinkTypeEnum.share }
}
);
jsonRes(res, {});
} catch (error) {
jsonRes(res, {
code: 500,
error
});
}
}

View File

@@ -1,446 +0,0 @@
// 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 { connectToDatabase, App } from '@/service/mongo';
import { FlowModuleTypeEnum, SpecialInputKeyEnum } from '@/constants/flow';
import { TaskResponseKeyEnum } from '@/constants/chat';
import { FlowInputItemType } from '@/types/flow';
const chatModelInput = ({
model,
temperature,
maxToken,
systemPrompt,
limitPrompt,
kbList
}: {
model: string;
temperature: number;
maxToken: number;
systemPrompt: string;
limitPrompt: string;
kbList: { kbId: string }[];
}): FlowInputItemType[] => [
{
key: 'model',
value: model,
type: 'custom',
label: '对话模型',
connected: true
},
{
key: 'temperature',
value: temperature,
label: '温度',
type: 'slider',
connected: true
},
{
key: 'maxToken',
value: maxToken,
type: 'custom',
label: '回复上限',
connected: true
},
{
key: 'systemPrompt',
value: systemPrompt,
type: 'textarea',
label: '系统提示词',
connected: true
},
{
key: 'limitPrompt',
label: '限定词',
type: 'textarea',
value: limitPrompt,
connected: true
},
{
key: 'switch',
type: 'target',
label: '触发器',
connected: kbList.length > 0
},
{
key: 'quoteQA',
type: 'target',
label: '引用内容',
connected: kbList.length > 0
},
{
key: 'history',
type: 'target',
label: '聊天记录',
connected: true
},
{
key: 'userChatInput',
type: 'target',
label: '用户问题',
connected: true
}
];
const chatTemplate = ({
model,
temperature,
maxToken,
systemPrompt,
limitPrompt
}: {
model: string;
temperature: number;
maxToken: number;
systemPrompt: string;
limitPrompt: string;
}) => {
return [
{
flowType: FlowModuleTypeEnum.questionInput,
inputs: [
{
key: 'userChatInput',
connected: true
}
],
outputs: [
{
key: 'userChatInput',
targets: [
{
moduleId: 'chatModule',
key: 'userChatInput'
}
]
}
],
position: {
x: 464.32198615344566,
y: 1602.2698463081606
},
moduleId: 'userChatInput'
},
{
flowType: FlowModuleTypeEnum.historyNode,
inputs: [
{
key: 'maxContext',
value: 10,
connected: true
},
{
key: 'history',
connected: true
}
],
outputs: [
{
key: 'history',
targets: [
{
moduleId: 'chatModule',
key: 'history'
}
]
}
],
position: {
x: 452.5466249541586,
y: 1276.3930310334215
},
moduleId: 'history'
},
{
flowType: FlowModuleTypeEnum.chatNode,
inputs: chatModelInput({
model,
temperature,
maxToken,
systemPrompt,
limitPrompt,
kbList: []
}),
outputs: [
{
key: TaskResponseKeyEnum.answerText,
targets: []
}
],
position: {
x: 981.9682828103937,
y: 890.014595014464
},
moduleId: 'chatModule'
}
];
};
const kbTemplate = ({
model,
temperature,
maxToken,
systemPrompt,
limitPrompt,
kbList = [],
searchSimilarity,
searchLimit,
searchEmptyText
}: {
model: string;
temperature: number;
maxToken: number;
systemPrompt: string;
limitPrompt: string;
kbList: { kbId: string }[];
searchSimilarity: number;
searchLimit: number;
searchEmptyText: string;
}) => {
return [
{
flowType: FlowModuleTypeEnum.questionInput,
inputs: [
{
key: 'userChatInput',
connected: true
}
],
outputs: [
{
key: 'userChatInput',
targets: [
{
moduleId: 'chatModule',
key: 'userChatInput'
},
{
moduleId: 'kbSearch',
key: 'userChatInput'
}
]
}
],
position: {
x: 464.32198615344566,
y: 1602.2698463081606
},
moduleId: 'userChatInput'
},
{
flowType: FlowModuleTypeEnum.historyNode,
inputs: [
{
key: 'maxContext',
value: 10,
connected: true
},
{
key: 'history',
connected: true
}
],
outputs: [
{
key: 'history',
targets: [
{
moduleId: 'chatModule',
key: 'history'
}
]
}
],
position: {
x: 452.5466249541586,
y: 1276.3930310334215
},
moduleId: 'history'
},
{
flowType: FlowModuleTypeEnum.kbSearchNode,
inputs: [
{
key: 'kbList',
value: kbList,
connected: true
},
{
key: 'similarity',
value: searchSimilarity,
connected: true
},
{
key: 'limit',
value: searchLimit,
connected: true
},
{
key: 'switch',
connected: false
},
{
key: 'userChatInput',
connected: true
}
],
outputs: [
{
key: 'isEmpty',
targets: searchEmptyText
? [
{
moduleId: 'emptyText',
key: 'switch'
}
]
: [
{
moduleId: 'chatModule',
key: 'switch'
}
]
},
{
key: 'unEmpty',
targets: [
{
moduleId: 'chatModule',
key: 'switch'
}
]
},
{
key: 'quoteQA',
targets: [
{
moduleId: 'chatModule',
key: 'quoteQA'
}
]
}
],
position: {
x: 956.0838440206068,
y: 887.462827870246
},
moduleId: 'kbSearch'
},
...(searchEmptyText
? [
{
flowType: FlowModuleTypeEnum.answerNode,
inputs: [
{
key: 'switch',
connected: true
},
{
key: SpecialInputKeyEnum.answerText,
value: searchEmptyText,
connected: true
}
],
outputs: [],
position: {
x: 1553.5815811529146,
y: 637.8753731306779
},
moduleId: 'emptyText'
}
]
: []),
{
flowType: FlowModuleTypeEnum.chatNode,
inputs: chatModelInput({ model, temperature, maxToken, systemPrompt, limitPrompt, kbList }),
outputs: [
{
key: TaskResponseKeyEnum.answerText,
targets: []
}
],
position: {
x: 1551.71405495818,
y: 977.4911578918461
},
moduleId: 'chatModule'
}
];
};
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
try {
await authUser({ req, authRoot: true });
await connectToDatabase();
const { limit = 1000 } = req.body as { limit: number };
let skip = 0;
const total = await App.countDocuments();
let promise = Promise.resolve();
console.log(total);
for (let i = 0; i < total; i += limit) {
const skipVal = skip;
skip += limit;
promise = promise
.then(() => init(limit, skipVal))
.then(() => {
console.log(skipVal);
});
}
await promise;
jsonRes(res, {});
} catch (error) {
jsonRes(res, {
code: 500,
error
});
}
}
async function init(limit: number, skip: number) {
// 遍历 app
const apps = await App.find(
{
chat: { $ne: null },
modules: { $exists: false }
// userId: '63f9a14228d2a688d8dc9e1b'
},
'_id chat'
).limit(limit);
return Promise.all(
apps.map(async (app) => {
if (!app.chat) return app;
const modules = (() => {
if (app.chat.relatedKbs.length === 0) {
return chatTemplate({
model: app.chat.chatModel,
temperature: app.chat.temperature,
maxToken: app.chat.maxToken,
systemPrompt: app.chat.systemPrompt,
limitPrompt: app.chat.limitPrompt
});
} else {
return kbTemplate({
model: app.chat.chatModel,
temperature: app.chat.temperature,
maxToken: app.chat.maxToken,
systemPrompt: app.chat.systemPrompt,
limitPrompt: app.chat.limitPrompt,
kbList: app.chat.relatedKbs.map((id) => ({ kbId: id })),
searchEmptyText: app.chat.searchEmptyText,
searchLimit: app.chat.searchLimit,
searchSimilarity: app.chat.searchSimilarity
});
}
})();
await App.findByIdAndUpdate(app.id, {
modules
});
return modules;
})
);
}

View File

@@ -1,44 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, Collection, App } from '@/service/mongo';
import { authUser } from '@/service/utils/auth';
/* 模型收藏切换 */
export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
const { appId } = req.query as { appId: string };
if (!appId) {
throw new Error('缺少参数');
}
// 凭证校验
const { userId } = await authUser({ req, authToken: true });
await connectToDatabase();
const collectionRecord = await Collection.findOne({
userId,
modelId: appId
});
if (collectionRecord) {
await Collection.findByIdAndRemove(collectionRecord._id);
} else {
await Collection.create({
userId,
modelId: appId
});
}
await App.findByIdAndUpdate(appId, {
'share.collection': await Collection.countDocuments({ modelId: appId })
});
jsonRes(res);
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}

View File

@@ -1,106 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, App } from '@/service/mongo';
import type { PagingData } from '@/types';
import type { ShareAppItem } from '@/types/app';
import { authUser } from '@/service/utils/auth';
import { Types } from 'mongoose';
/* 获取模型列表 */
export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
const {
searchText = '',
pageNum = 1,
pageSize = 20
} = req.body as { searchText: string; pageNum: number; pageSize: number };
await connectToDatabase();
const { userId } = await authUser({ req, authToken: true });
const regex = new RegExp(searchText, 'i');
const where = {
$and: [
{ 'share.isShare': true },
{
$or: [{ name: { $regex: regex } }, { intro: { $regex: regex } }]
}
]
};
const pipeline = [
{
$match: where
},
{
$lookup: {
from: 'collections',
let: { modelId: '$_id' },
pipeline: [
{
$match: {
$expr: {
$and: [
{ $eq: ['$modelId', '$$modelId'] },
{
$eq: ['$userId', userId ? new Types.ObjectId(userId) : new Types.ObjectId()]
}
]
}
}
}
],
as: 'collections'
}
},
{
$project: {
_id: 1,
avatar: { $ifNull: ['$avatar', '/icon/logo.svg'] },
name: 1,
userId: 1,
intro: 1,
share: 1,
isCollection: {
$cond: {
if: { $gt: [{ $size: '$collections' }, 0] },
then: true,
else: false
}
}
}
},
{
$sort: { 'share.topNum': -1, 'share.collection': -1 }
},
{
$skip: (pageNum - 1) * pageSize
},
{
$limit: pageSize
}
];
// 获取被分享的模型
const [models, total] = await Promise.all([
// @ts-ignore
App.aggregate(pipeline),
App.countDocuments(where)
]);
jsonRes<PagingData<ShareAppItem>>(res, {
data: {
pageNum,
pageSize,
data: models,
total
}
});
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}

View File

@@ -1,37 +0,0 @@
// Next.js API route support: https://nextjs.org/docs/api-routes/introduction
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, OpenApi } from '@/service/mongo';
import { authUser } from '@/service/utils/auth';
import { UserOpenApiKey } from '@/types/openapi';
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
try {
const { userId } = await authUser({ req, authToken: true });
await connectToDatabase();
const findResponse = await OpenApi.find({ userId }).sort({ _id: -1 });
// jus save four data
const apiKeys = findResponse.map<UserOpenApiKey>(
({ _id, apiKey, createTime, lastUsedTime }) => {
return {
id: _id,
apiKey: `******${apiKey.substring(apiKey.length - 4)}`,
createTime,
lastUsedTime
};
}
);
jsonRes(res, {
data: apiKeys
});
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}

View File

@@ -1,175 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, TrainingData, KB } from '@/service/mongo';
import { authUser } from '@/service/utils/auth';
import { authKb } from '@/service/utils/auth';
import { withNextCors } from '@/service/utils/tools';
import { PgTrainingTableName, TrainingModeEnum } from '@/constants/plugin';
import { startQueue } from '@/service/utils/tools';
import { PgClient } from '@/service/pg';
import { modelToolMap } from '@/utils/plugin';
export type DateItemType = { a: string; q: string; source?: string };
export type Props = {
kbId: string;
data: DateItemType[];
mode: `${TrainingModeEnum}`;
prompt?: string;
};
export type Response = {
insertLen: number;
};
const modeMaxToken = {
[TrainingModeEnum.index]: 6000,
[TrainingModeEnum.qa]: 12000
};
export default withNextCors(async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
const { kbId, data, mode, prompt } = req.body as Props;
if (!kbId || !Array.isArray(data)) {
throw new Error('缺少参数');
}
await connectToDatabase();
// 凭证校验
const { userId } = await authUser({ req });
jsonRes<Response>(res, {
data: await pushDataToKb({
kbId,
data,
userId,
mode,
prompt
})
});
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
});
export async function pushDataToKb({
userId,
kbId,
data,
mode,
prompt
}: { userId: string } & Props): Promise<Response> {
await authKb({
userId,
kbId
});
// 过滤重复的 qa 内容
const set = new Set();
const filterData: DateItemType[] = [];
data.forEach((item) => {
const text = item.q + item.a;
// count token
const token = modelToolMap.countTokens({
model: 'gpt-3.5-turbo',
messages: [{ obj: 'System', value: item.q }]
});
if (token > modeMaxToken[TrainingModeEnum.qa]) {
return;
}
if (!set.has(text)) {
filterData.push(item);
set.add(text);
}
});
// 数据库去重
const insertData = (
await Promise.allSettled(
filterData.map(async ({ q, a = '', source }) => {
if (mode !== TrainingModeEnum.index) {
return Promise.resolve({
q,
a,
source
});
}
if (!q) {
return Promise.reject('q为空');
}
q = q.replace(/\\n/g, '\n').trim().replace(/'/g, '"');
a = a.replace(/\\n/g, '\n').trim().replace(/'/g, '"');
// Exactly the same data, not push
try {
const { rows } = await PgClient.query(`
SELECT COUNT(*) > 0 AS exists
FROM ${PgTrainingTableName}
WHERE md5(q)=md5('${q}') AND md5(a)=md5('${a}') AND user_id='${userId}' AND kb_id='${kbId}'
`);
const exists = rows[0]?.exists || false;
if (exists) {
return Promise.reject('已经存在');
}
} catch (error) {
console.log(error);
error;
}
return Promise.resolve({
q,
a,
source
});
})
)
)
.filter((item) => item.status === 'fulfilled')
.map<DateItemType>((item: any) => item.value);
const vectorModel = await (async () => {
if (mode === TrainingModeEnum.index) {
return (await KB.findById(kbId, 'vectorModel'))?.vectorModel || global.vectorModels[0].model;
}
return global.vectorModels[0].model;
})();
// 插入记录
await TrainingData.insertMany(
insertData.map((item) => ({
q: item.q,
a: item.a,
source: item.source,
userId,
kbId,
mode,
prompt,
vectorModel
}))
);
insertData.length > 0 && startQueue();
return {
insertLen: insertData.length
};
}
export const config = {
api: {
bodyParser: {
sizeLimit: '20mb'
}
}
};

View File

@@ -1,60 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { authUser } from '@/service/utils/auth';
import { PgClient } from '@/service/pg';
import { withNextCors } from '@/service/utils/tools';
import { getVector } from '../plugin/vector';
import type { KbTestItemType } from '@/types/plugin';
import { PgTrainingTableName } from '@/constants/plugin';
import { KB } from '@/service/mongo';
export type Props = {
kbId: string;
text: string;
};
export type Response = KbTestItemType['results'];
export default withNextCors(async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
const { kbId, text } = req.body as Props;
if (!kbId || !text) {
throw new Error('缺少参数');
}
// 凭证校验
const [{ userId }, kb] = await Promise.all([
authUser({ req }),
KB.findById(kbId, 'vectorModel')
]);
if (!userId || !kb) {
throw new Error('缺少用户ID');
}
const { vectors } = await getVector({
model: kb.vectorModel,
userId,
input: [text]
});
const response: any = await PgClient.query(
`BEGIN;
SET LOCAL ivfflat.probes = ${global.systemEnv.pgIvfflatProbe || 10};
select id,q,a,source,(vector <#> '[${
vectors[0]
}]') * -1 AS score from ${PgTrainingTableName} where kb_id='${kbId}' AND user_id='${userId}' order by vector <#> '[${
vectors[0]
}]' limit 12;
COMMIT;`
);
jsonRes<Response>(res, { data: response?.[2]?.rows || [] });
} catch (err) {
console.log(err);
jsonRes(res, {
code: 500,
error: err
});
}
});

View File

@@ -1,71 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { authUser } from '@/service/utils/auth';
import { PgClient } from '@/service/pg';
import { withNextCors } from '@/service/utils/tools';
import { KB, connectToDatabase } from '@/service/mongo';
import { getVector } from '../plugin/vector';
import { PgTrainingTableName } from '@/constants/plugin';
export type Props = {
dataId: string;
kbId: string;
a?: string;
q?: string;
};
export default withNextCors(async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
const { dataId, a = '', q = '', kbId } = req.body as Props;
if (!dataId) {
throw new Error('缺少参数');
}
await connectToDatabase();
// auth user and get kb
const [{ userId }, kb] = await Promise.all([
authUser({ req }),
KB.findById(kbId, 'vectorModel')
]);
if (!kb) {
throw new Error("Can't find database");
}
// get vector
const { vectors = [] } = await (async () => {
if (q) {
return getVector({
userId,
input: [q],
model: kb.vectorModel
});
}
return { vectors: [[]] };
})();
// 更新 pg 内容.仅修改a不需要更新向量。
await PgClient.update(PgTrainingTableName, {
where: [['id', dataId], 'AND', ['user_id', userId]],
values: [
{ key: 'source', value: '手动修改' },
{ key: 'a', value: a.replace(/'/g, '"') },
...(q
? [
{ key: 'q', value: q.replace(/'/g, '"') },
{ key: 'vector', value: `[${vectors[0]}]` }
]
: [])
]
});
jsonRes(res);
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
});

View File

@@ -1,97 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { authBalanceByUid, authUser } from '@/service/utils/auth';
import { withNextCors } from '@/service/utils/tools';
import { getAIChatApi, axiosConfig } from '@/service/ai/openai';
import { pushGenerateVectorBill } from '@/service/events/pushBill';
type Props = {
model: string;
input: string[];
};
type Response = {
tokenLen: number;
vectors: number[][];
};
export default withNextCors(async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
const { userId } = await authUser({ req });
let { input, model } = req.query as Props;
if (!Array.isArray(input)) {
throw new Error('缺少参数');
}
jsonRes<Response>(res, {
data: await getVector({ userId, input, model })
});
} catch (err) {
console.log(err);
jsonRes(res, {
code: 500,
error: err
});
}
});
export async function getVector({
model = 'text-embedding-ada-002',
userId,
input
}: { userId?: string } & Props) {
userId && (await authBalanceByUid(userId));
for (let i = 0; i < input.length; i++) {
if (!input[i]) {
return Promise.reject({
code: 500,
message: '向量生成模块输入内容为空'
});
}
}
// 获取 chatAPI
const chatAPI = getAIChatApi();
// 把输入的内容转成向量
const result = await chatAPI
.createEmbedding(
{
model,
input
},
{
timeout: 60000,
...axiosConfig()
}
)
.then((res) => {
if (!res.data?.data?.[0]?.embedding) {
// @ts-ignore
return Promise.reject(res.data?.error?.message || 'Embedding Error');
}
return {
tokenLen: res.data.usage.total_tokens || 0,
vectors: res.data.data.map((item) => unityDimensional(item.embedding))
};
});
userId &&
pushGenerateVectorBill({
userId,
tokenLen: result.tokenLen,
model
});
return result;
}
function unityDimensional(vector: number[]) {
let resultVector = vector;
const vectorLen = vector.length;
const zeroVector = new Array(1536 - vectorLen).fill(0);
return resultVector.concat(zeroVector);
}

View File

@@ -1,37 +0,0 @@
// Next.js API route support: https://nextjs.org/docs/api-routes/introduction
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, OpenApi } from '@/service/mongo';
import { authUser } from '@/service/utils/auth';
import { customAlphabet } from 'nanoid';
const nanoid = customAlphabet('abcdefghijklmnopqrstuvwxyz1234567890', 24);
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
try {
const { userId } = await authUser({ req, authToken: true });
await connectToDatabase();
const count = await OpenApi.find({ userId }).countDocuments();
if (count >= 10) {
throw new Error('最多 10 组 API 秘钥');
}
const apiKey = `fastgpt-${nanoid()}`;
await OpenApi.create({
userId,
apiKey
});
jsonRes(res, {
data: apiKey
});
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}

View File

@@ -1,67 +0,0 @@
// 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 type { ChatItemType } from '@/types/chat';
import { countOpenAIToken } from '@/utils/plugin/openai';
import { OpenAiChatEnum } from '@/constants/model';
type ModelType = `${OpenAiChatEnum}`;
type Props = {
messages: ChatItemType[];
model: ModelType;
maxLen: number;
};
type Response = ChatItemType[];
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
try {
await authUser({ req });
const { messages, model, maxLen } = req.body as Props;
if (!Array.isArray(messages) || !model || !maxLen) {
throw new Error('params is error');
}
return jsonRes<Response>(res, {
data: gpt_chatItemTokenSlice({
messages,
model,
maxToken: maxLen
})
});
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}
export function gpt_chatItemTokenSlice({
messages,
model = 'gpt-3.5-turbo',
maxToken
}: {
messages: ChatItemType[];
model?: string;
maxToken: number;
}) {
let result: ChatItemType[] = [];
for (let i = 0; i < messages.length; i++) {
const msgs = [...result, messages[i]];
const tokens = countOpenAIToken({ messages: msgs, model });
if (tokens < maxToken) {
result = msgs;
} else {
break;
}
}
return result.length === 0 && messages[0] ? [messages[0]] : result;
}

View File

@@ -1,51 +0,0 @@
// 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 axios from 'axios';
import { axiosConfig } from '@/service/ai/openai';
export type Props = {
input: string;
};
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
try {
await authUser({ req });
const result = await sensitiveCheck(req.body);
jsonRes(res, {
data: result,
message: result
});
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}
export async function sensitiveCheck({ input }: Props) {
const response = await axios({
...axiosConfig(),
method: 'POST',
url: `/moderations`,
data: {
input
}
});
const data = (response.data.results?.[0]?.category_scores as Record<string, number>) || {};
const values = Object.values(data);
for (const val of values) {
if (val > 0.2) {
return Promise.reject('您的内容不合规');
}
}
return '';
}

View File

@@ -1,35 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, KB } from '@/service/mongo';
import { authUser } from '@/service/utils/auth';
import type { CreateKbParams } from '@/api/request/kb';
export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
const { name, tags, avatar, vectorModel } = req.body as CreateKbParams;
if (!name || !vectorModel) {
throw new Error('缺少参数');
}
// 凭证校验
const { userId } = await authUser({ req, authToken: true });
await connectToDatabase();
const { _id } = await KB.create({
name,
userId,
tags,
vectorModel,
avatar
});
jsonRes(res, { data: _id });
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}

View File

@@ -1,84 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, User } from '@/service/mongo';
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<any>) {
try {
let { kbId } = req.query as {
kbId: string;
};
if (!kbId) {
throw new Error('缺少参数');
}
await connectToDatabase();
// 凭证校验
const { userId } = await authUser({ req, authToken: true });
const thirtyMinutesAgo = new Date(Date.now() - 30 * 60 * 1000);
// auth export times
const authTimes = await User.findOne(
{
_id: userId,
$or: [
{ 'limit.exportKbTime': { $exists: false } },
{ 'limit.exportKbTime': { $lte: thirtyMinutesAgo } }
]
},
'_id limit'
);
if (!authTimes) {
throw new Error('上次导出未到半小时,每半小时仅可导出一次。');
}
// 统计数据
const count = await PgClient.count(PgTrainingTableName, {
where: [['kb_id', kbId], 'AND', ['user_id', userId]]
});
// 从 pg 中获取所有数据
const pgData = await PgClient.select<{ q: string; a: string; source: string }>(
PgTrainingTableName,
{
where: [['kb_id', kbId], 'AND', ['user_id', userId]],
fields: ['q', 'a', 'source'],
order: [{ field: 'id', mode: 'DESC' }],
limit: count
}
);
const data: [string, string, string][] = pgData.rows.map((item) => [
item.q.replace(/\n/g, '\\n'),
item.a.replace(/\n/g, '\\n'),
item.source
]);
// update export time
await User.findByIdAndUpdate(userId, {
'limit.exportKbTime': new Date()
});
jsonRes(res, {
data
});
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}
export const config = {
api: {
bodyParser: {
sizeLimit: '100mb'
}
}
};

View File

@@ -1,47 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase } from '@/service/mongo';
import { authUser } from '@/service/utils/auth';
import { PgClient } from '@/service/pg';
import type { KbDataItemType } from '@/types/plugin';
import { PgTrainingTableName } from '@/constants/plugin';
export type Response = {
id: string;
q: string;
a: string;
source: string;
};
export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
let { dataId } = req.query as {
dataId: string;
};
if (!dataId) {
throw new Error('缺少参数');
}
// 凭证校验
const { userId } = await authUser({ req, authToken: true });
await connectToDatabase();
const where: any = [['user_id', userId], 'AND', ['id', dataId]];
const searchRes = await PgClient.select<KbDataItemType>(PgTrainingTableName, {
fields: ['kb_id', 'id', 'q', 'a', 'source'],
where,
limit: 1
});
jsonRes(res, {
data: searchRes.rows[0]
});
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}

View File

@@ -1,52 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, TrainingData } from '@/service/mongo';
import { authUser } from '@/service/utils/auth';
import { TrainingModeEnum } from '@/constants/plugin';
import { Types } from 'mongoose';
import { startQueue } from '@/service/utils/tools';
/* 拆分数据成QA */
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
try {
const { kbId, init = false } = req.body as { kbId: string; init: boolean };
if (!kbId) {
throw new Error('参数错误');
}
await connectToDatabase();
const { userId } = await authUser({ req, authToken: true });
// split queue data
const result = await TrainingData.aggregate([
{
$match: {
userId: new Types.ObjectId(userId),
kbId: new Types.ObjectId(kbId)
}
},
{
$group: {
_id: '$mode',
count: { $sum: 1 }
}
}
]);
jsonRes(res, {
data: {
qaListLen: result.find((item) => item._id === TrainingModeEnum.qa)?.count || 0,
vectorListLen: result.find((item) => item._id === TrainingModeEnum.index)?.count || 0
}
});
if (init) {
startQueue();
}
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}

View File

@@ -1,88 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, KB } from '@/service/mongo';
import { authKb, authUser } from '@/service/utils/auth';
import { withNextCors } from '@/service/utils/tools';
import { PgTrainingTableName } from '@/constants/plugin';
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';
export type Props = {
kbId: string;
data: { a: string; q: string; source?: string };
};
export default withNextCors(async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
await connectToDatabase();
const { kbId, data = { q: '', a: '' } } = req.body as Props;
if (!kbId || !data?.q) {
throw new Error('缺少参数');
}
// 凭证校验
const { userId } = await authUser({ req });
// auth kb
const kb = await authKb({ kbId, userId });
const q = data?.q?.replace(/\\n/g, '\n').trim().replace(/'/g, '"');
const a = data?.a?.replace(/\\n/g, '\n').trim().replace(/'/g, '"');
// token check
const token = modelToolMap.countTokens({
model: 'gpt-3.5-turbo',
messages: [{ obj: 'System', value: q }]
});
if (token > getVectorModel(kb.vectorModel).maxToken) {
throw new Error('Over Tokens');
}
const { rows: existsRows } = await PgClient.query(`
SELECT COUNT(*) > 0 AS exists
FROM ${PgTrainingTableName}
WHERE md5(q)=md5('${q}') AND md5(a)=md5('${a}') AND user_id='${userId}' AND kb_id='${kbId}'
`);
const exists = existsRows[0]?.exists || false;
if (exists) {
throw new Error('已经存在完全一致的数据');
}
const { vectors } = await getVector({
model: kb.vectorModel,
input: [q],
userId
});
const response = await insertKbItem({
userId,
kbId,
data: [
{
q,
a,
source: data.source,
vector: vectors[0]
}
]
});
// @ts-ignore
const id = response?.rows?.[0]?.id || '';
jsonRes(res, {
data: id
});
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
});

View File

@@ -1,56 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, KB, App, TrainingData } from '@/service/mongo';
import { authUser } from '@/service/utils/auth';
import { PgClient } from '@/service/pg';
import { Types } from 'mongoose';
import { PgTrainingTableName } from '@/constants/plugin';
export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
const { id } = req.query as {
id: string;
};
if (!id) {
throw new Error('缺少参数');
}
// 凭证校验
const { userId } = await authUser({ req, authToken: true });
await connectToDatabase();
// delete all pg data
await PgClient.delete(PgTrainingTableName, {
where: [['user_id', userId], 'AND', ['kb_id', id]]
});
// delete training data
await TrainingData.deleteMany({
userId,
kbId: id
});
// delete related app
await App.updateMany(
{
userId
},
{ $pull: { 'chat.relatedKbs': new Types.ObjectId(id) } }
);
// delete kb data
await KB.findOneAndDelete({
_id: id,
userId
});
jsonRes(res);
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}

View File

@@ -1,41 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, KB } from '@/service/mongo';
import { authUser } from '@/service/utils/auth';
import { KbListItemType } from '@/types/plugin';
import { getVectorModel } from '@/service/utils/data';
export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
// 凭证校验
const { userId } = await authUser({ req, authToken: true });
await connectToDatabase();
const kbList = await KB.find(
{
userId
},
'_id avatar name tags vectorModel'
).sort({ updateTime: -1 });
const data = await Promise.all(
kbList.map(async (item) => ({
_id: item._id,
avatar: item.avatar,
name: item.name,
tags: item.tags,
vectorModel: getVectorModel(item.vectorModel)
}))
);
jsonRes<KbListItemType[]>(res, {
data
});
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}

View File

@@ -1,39 +0,0 @@
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { connectToDatabase, KB } from '@/service/mongo';
import { authUser } from '@/service/utils/auth';
import type { KbUpdateParams } from '@/api/request/kb';
export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
const { id, name, tags, avatar } = req.body as KbUpdateParams;
if (!id || !name) {
throw new Error('缺少参数');
}
// 凭证校验
const { userId } = await authUser({ req, authToken: true });
await connectToDatabase();
await KB.findOneAndUpdate(
{
_id: id,
userId
},
{
avatar,
name,
tags: tags.split(' ').filter((item) => item)
}
);
jsonRes(res);
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}

View File

@@ -1,113 +0,0 @@
import type { FeConfigsType } from '@/types';
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { readFileSync } from 'fs';
import {
type QAModelItemType,
type ChatModelItemType,
type VectorModelItemType
} from '@/types/model';
export type InitDateResponse = {
chatModels: ChatModelItemType[];
qaModel: QAModelItemType;
vectorModels: VectorModelItemType[];
feConfigs: FeConfigsType;
};
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
if (!global.feConfigs) {
await getInitConfig();
}
jsonRes<InitDateResponse>(res, {
data: {
feConfigs: global.feConfigs,
chatModels: global.chatModels,
qaModel: global.qaModel,
vectorModels: global.vectorModels
}
});
}
const defaultSystemEnv = {
vectorMaxProcess: 15,
qaMaxProcess: 15,
pgIvfflatProbe: 20
};
const defaultFeConfigs = {
show_emptyChat: true,
show_register: false,
show_appStore: false,
show_userDetail: false,
show_git: true,
systemTitle: 'FastGPT',
authorText: 'Made by FastGPT Team.'
};
const defaultChatModels = [
{
model: 'gpt-3.5-turbo',
name: 'GPT35-4k',
contextMaxToken: 4000,
quoteMaxToken: 2400,
maxTemperature: 1.2,
price: 0
},
{
model: 'gpt-3.5-turbo-16k',
name: 'GPT35-16k',
contextMaxToken: 16000,
quoteMaxToken: 8000,
maxTemperature: 1.2,
price: 0
},
{
model: 'gpt-4',
name: 'GPT4-8k',
contextMaxToken: 8000,
quoteMaxToken: 4000,
maxTemperature: 1.2,
price: 0
}
];
const defaultQAModel = {
model: 'gpt-3.5-turbo-16k',
name: 'GPT35-16k',
maxToken: 16000,
price: 0
};
const defaultVectorModels: VectorModelItemType[] = [
{
model: 'text-embedding-ada-002',
name: 'Embedding-2',
price: 0,
defaultToken: 500,
maxToken: 3000
}
];
export async function getInitConfig() {
try {
const filename =
process.env.NODE_ENV === 'development' ? 'data/config.local.json' : '/app/data/config.json';
const res = JSON.parse(readFileSync(filename, 'utf-8'));
console.log(res);
global.systemEnv = res.SystemParams || defaultSystemEnv;
global.feConfigs = res.FeConfig || defaultFeConfigs;
global.chatModels = res.ChatModels || defaultChatModels;
global.qaModel = res.QAModel || defaultQAModel;
global.vectorModels = res.VectorModels || defaultVectorModels;
} catch (error) {
setDefaultData();
console.log('get init config error, set default', error);
}
}
export function setDefaultData() {
global.systemEnv = defaultSystemEnv;
global.feConfigs = defaultFeConfigs;
global.chatModels = defaultChatModels;
global.qaModel = defaultQAModel;
global.vectorModels = defaultVectorModels;
}

View File

@@ -1,120 +0,0 @@
// Next.js API route support: https://nextjs.org/docs/api-routes/introduction
import type { NextApiRequest, NextApiResponse } from 'next';
import { jsonRes } from '@/service/response';
import { User } from '@/service/models/user';
import { generateToken, setCookie } from '@/service/utils/tools';
import axios from 'axios';
import { parseQueryString } from '@/utils/tools';
import { customAlphabet } from 'nanoid';
const nanoid = customAlphabet('abcdefghijklmnopqrstuvwxyz1234567890', 8);
type GithubAccessTokenType = {
access_token: string;
expires_in: number;
refresh_token: string;
refresh_token_expires_in: number;
token_type: 'bearer';
scope: string;
};
type GithubUserType = {
login: string;
email: string;
avatar_url: string;
};
export default async function handler(req: NextApiRequest, res: NextApiResponse<any>) {
try {
const { code, inviterId } = req.query as { code: string; inviterId?: string };
const { data: gitAccessToken } = await axios.post<string>(
`https://github.com/login/oauth/access_token?client_id=${global.feConfigs.gitLoginKey}&client_secret=${global.systemEnv.gitLoginSecret}&code=${code}`
);
const jsonGitAccessToken = parseQueryString(gitAccessToken) as GithubAccessTokenType;
const access_token = jsonGitAccessToken?.access_token;
if (!access_token) {
throw new Error('access_token is null');
}
const { data } = await axios.get<GithubUserType>('https://api.github.com/user', {
headers: {
Authorization: `Bearer ${access_token}`
}
});
const { login, avatar_url } = data;
const username = `git-${login}`;
try {
jsonRes(res, {
data: await loginByUsername({ username, res })
});
} catch (err: any) {
if (err?.code === 500) {
jsonRes(res, {
data: await registerUser({ username, avatar: avatar_url, res, inviterId })
});
return;
}
throw new Error(err);
}
} catch (err) {
jsonRes(res, {
code: 500,
error: err
});
}
}
export async function loginByUsername({
username,
res
}: {
username: string;
res: NextApiResponse;
}) {
const user = await User.findOne({ username });
if (!user) {
return Promise.reject({
code: 500
});
}
const token = generateToken(user._id);
setCookie(res, token);
return { user, token };
}
export async function registerUser({
username,
avatar,
inviterId,
res
}: {
username: string;
avatar?: string;
inviterId?: string;
res: NextApiResponse;
}) {
const response = await User.create({
username,
avatar,
password: nanoid(),
inviterId: inviterId ? inviterId : undefined
});
// 根据 id 获取用户信息
const user = await User.findById(response._id);
if (!user) {
return Promise.reject('获取用户信息异常');
}
const token = generateToken(user._id);
setCookie(res, token);
return {
user,
token
};
}

View File

@@ -1,92 +0,0 @@
import React, { useEffect, useState } from 'react';
import { Box, Divider, Flex, useTheme, Button, Skeleton, useDisclosure } from '@chakra-ui/react';
import { useCopyData } from '@/utils/tools';
import dynamic from 'next/dynamic';
import MyIcon from '@/components/Icon';
import { useGlobalStore } from '@/store/global';
const APIKeyModal = dynamic(() => import('@/components/APIKeyModal'), {
ssr: false
});
const API = ({ appId }: { appId: string }) => {
const theme = useTheme();
const { copyData } = useCopyData();
const [baseUrl, setBaseUrl] = useState('https://fastgpt.run/api/openapi');
const {
isOpen: isOpenAPIModal,
onOpen: onOpenAPIModal,
onClose: onCloseAPIModal
} = useDisclosure();
const [isLoaded, setIsLoaded] = useState(false);
const { isPc } = useGlobalStore();
useEffect(() => {
setBaseUrl(`${location.origin}/api/openapi`);
}, []);
return (
<Flex flexDirection={'column'} pt={[0, 5]} h={'100%'}>
<Flex px={5} alignItems={'center'}>
<Box flex={1}>
AppId:
<Box
as={'span'}
ml={2}
fontWeight={'bold'}
cursor={'pointer'}
onClick={() => copyData(appId, '已复制 AppId')}
>
{appId}
</Box>
</Box>
{isPc && (
<>
<Flex
bg={'myWhite.600'}
py={2}
px={4}
borderRadius={'md'}
cursor={'pointer'}
onClick={() => copyData(baseUrl, '已复制 API 地址')}
>
<Box border={theme.borders.md} px={2} borderRadius={'md'} fontSize={'sm'}>
API服务器
</Box>
<Box ml={2} color={'myGray.900'} fontSize={['sm', 'md']}>
{baseUrl}
</Box>
</Flex>
<Button
ml={3}
leftIcon={<MyIcon name={'apikey'} w={'16px'} color={''} />}
variant={'base'}
onClick={onOpenAPIModal}
>
API
</Button>
</>
)}
</Flex>
<Divider mt={3} />
<Box flex={'1 0 0'} h={0}>
<Skeleton h="100%" isLoaded={isLoaded} fadeDuration={2}>
<iframe
style={{
width: '100%',
height: '100%'
}}
src="https://kjqvjse66l.feishu.cn/docx/DmLedTWtUoNGX8xui9ocdUEjnNh"
frameBorder="0"
onLoad={() => setIsLoaded(true)}
onError={() => setIsLoaded(true)}
/>
</Skeleton>
</Box>
{isOpenAPIModal && <APIKeyModal onClose={onCloseAPIModal} />}
</Flex>
);
};
export default API;

View File

@@ -1,138 +0,0 @@
import React from 'react';
import { NodeProps } from 'reactflow';
import { Box, Input, Button, Flex } from '@chakra-ui/react';
import NodeCard from '../modules/NodeCard';
import { FlowModuleItemType } from '@/types/flow';
import Divider from '../modules/Divider';
import Container from '../modules/Container';
import RenderInput from '../render/RenderInput';
import type { ClassifyQuestionAgentItemType } from '@/types/app';
import { customAlphabet } from 'nanoid';
const nanoid = customAlphabet('abcdefghijklmnopqrstuvwxyz1234567890', 4);
import MyIcon from '@/components/Icon';
import { FlowOutputItemTypeEnum, FlowValueTypeEnum, SpecialInputKeyEnum } from '@/constants/flow';
import SourceHandle from '../render/SourceHandle';
const NodeCQNode = ({ data }: NodeProps<FlowModuleItemType>) => {
const { moduleId, inputs, outputs, onChangeNode } = data;
return (
<NodeCard minW={'400px'} {...data}>
<Divider text="Input" />
<Container>
<RenderInput
moduleId={moduleId}
onChangeNode={onChangeNode}
flowInputList={inputs}
CustomComponent={{
[SpecialInputKeyEnum.agents]: ({
key: agentKey,
value: agents = [],
...props
}: {
key: string;
value?: ClassifyQuestionAgentItemType[];
}) => (
<Box>
{agents.map((item, i) => (
<Flex key={item.key} mb={4} alignItems={'center'}>
<MyIcon
mr={2}
name={'minus'}
w={'14px'}
cursor={'pointer'}
color={'myGray.600'}
_hover={{ color: 'myGray.900' }}
onClick={() => {
const newInputValue = agents.filter((input) => input.key !== item.key);
const newOutputVal = outputs.filter((output) => output.key !== item.key);
onChangeNode({
moduleId,
type: 'inputs',
key: agentKey,
value: {
...props,
key: agentKey,
value: newInputValue
}
});
onChangeNode({
moduleId,
type: 'outputs',
key: '',
value: newOutputVal
});
}}
/>
<Box flex={1}>
<Box flex={1}>{i + 1}</Box>
<Box position={'relative'}>
<Input
mt={1}
defaultValue={item.value}
onChange={(e) => {
const newVal = agents.map((val) =>
val.key === item.key
? {
...val,
value: e.target.value
}
: val
);
onChangeNode({
moduleId,
type: 'inputs',
key: agentKey,
value: {
...props,
key: agentKey,
value: newVal
}
});
}}
/>
<SourceHandle handleKey={item.key} valueType={FlowValueTypeEnum.boolean} />
</Box>
</Box>
</Flex>
))}
<Button
onClick={() => {
const key = nanoid();
const newInputValue = agents.concat({ value: '', key });
const newOutputValue = outputs.concat({
key,
label: '',
type: FlowOutputItemTypeEnum.hidden,
targets: []
});
onChangeNode({
moduleId,
type: 'inputs',
key: agentKey,
value: {
...props,
key: agentKey,
value: newInputValue
}
});
onChangeNode({
moduleId,
type: 'outputs',
key: agentKey,
value: newOutputValue
});
}}
>
</Button>
</Box>
)
}}
/>
</Container>
</NodeCard>
);
};
export default React.memo(NodeCQNode);

View File

@@ -1,131 +0,0 @@
import React, { useMemo } from 'react';
import { NodeProps } from 'reactflow';
import NodeCard from '../modules/NodeCard';
import { FlowModuleItemType } from '@/types/flow';
import Divider from '../modules/Divider';
import Container from '../modules/Container';
import RenderInput from '../render/RenderInput';
import RenderOutput from '../render/RenderOutput';
import { FlowOutputItemTypeEnum } from '@/constants/flow';
import MySelect from '@/components/Select';
import { chatModelList } from '@/store/static';
import MySlider from '@/components/Slider';
import { Box } from '@chakra-ui/react';
import { formatPrice } from '@/utils/user';
const NodeChat = ({ data }: NodeProps<FlowModuleItemType>) => {
const { moduleId, inputs, outputs, onChangeNode } = data;
const outputsLen = useMemo(
() => outputs.filter((item) => item.type !== FlowOutputItemTypeEnum.hidden).length,
[outputs]
);
return (
<NodeCard minW={'400px'} {...data}>
<Divider text="Input" />
<Container>
<RenderInput
moduleId={moduleId}
onChangeNode={onChangeNode}
flowInputList={inputs}
CustomComponent={{
model: (inputItem) => {
const list = chatModelList.map((item) => {
const priceStr = `(${formatPrice(item.price, 1000)}元/1k Tokens)`;
return {
value: item.model,
label: `${item.name}${priceStr}`
};
});
return (
<MySelect
width={'100%'}
value={inputItem.value}
list={list}
onchange={(e) => {
onChangeNode({
moduleId,
type: 'inputs',
key: inputItem.key,
value: {
...inputItem,
value: e
}
});
// update max tokens
const model =
chatModelList.find((item) => item.model === e) || chatModelList[0];
if (!model) return;
onChangeNode({
moduleId,
type: 'inputs',
key: 'maxToken',
value: {
...inputs.find((input) => input.key === 'maxToken'),
markList: [
{ label: '100', value: 100 },
{ label: `${model.contextMaxToken}`, value: model.contextMaxToken }
],
max: model.contextMaxToken,
value: model.contextMaxToken / 2
}
});
}}
/>
);
},
maxToken: (inputItem) => {
const model = inputs.find((item) => item.key === 'model')?.value;
const modelData = chatModelList.find((item) => item.model === model);
const maxToken = modelData ? modelData.contextMaxToken : 4000;
const markList = [
{ label: '100', value: 100 },
{ label: `${maxToken}`, value: maxToken }
];
return (
<Box pt={5} pb={4} px={2}>
<MySlider
markList={markList}
width={'100%'}
min={inputItem.min || 100}
max={maxToken}
step={inputItem.step || 1}
value={inputItem.value}
onChange={(e) => {
onChangeNode({
moduleId,
type: 'inputs',
key: inputItem.key,
value: {
...inputItem,
value: e
}
});
}}
/>
</Box>
);
}
}}
/>
</Container>
{outputsLen > 0 && (
<>
<Divider text="Output" />
<Container>
<RenderOutput
onChangeNode={onChangeNode}
moduleId={moduleId}
flowOutputList={outputs}
/>
</Container>
</>
)}
</NodeCard>
);
};
export default React.memo(NodeChat);

View File

@@ -1,110 +0,0 @@
import React, { useMemo, useState } from 'react';
import { NodeProps } from 'reactflow';
import { FlowModuleItemType } from '@/types/flow';
import { Flex, Box, Button, useTheme, useDisclosure, Grid } from '@chakra-ui/react';
import { useUserStore } from '@/store/user';
import { useQuery } from '@tanstack/react-query';
import NodeCard from '../modules/NodeCard';
import Divider from '../modules/Divider';
import Container from '../modules/Container';
import RenderInput from '../render/RenderInput';
import RenderOutput from '../render/RenderOutput';
import { KBSelectModal } from '../../../KBSelectModal';
import type { SelectedKbType } from '@/types/plugin';
import Avatar from '@/components/Avatar';
const KBSelect = ({
activeKbs = [],
onChange
}: {
activeKbs: SelectedKbType;
onChange: (e: SelectedKbType) => void;
}) => {
const theme = useTheme();
const { myKbList, loadKbList } = useUserStore();
const {
isOpen: isOpenKbSelect,
onOpen: onOpenKbSelect,
onClose: onCloseKbSelect
} = useDisclosure();
const showKbList = useMemo(
() => myKbList.filter((item) => activeKbs.find((kb) => kb.kbId === item._id)),
[myKbList, activeKbs]
);
useQuery(['initkb'], loadKbList);
return (
<>
<Grid gridTemplateColumns={'1fr 1fr'} gridGap={4}>
<Button h={'36px'} onClick={onOpenKbSelect}>
</Button>
{showKbList.map((item) => (
<Flex
key={item._id}
alignItems={'center'}
h={'36px'}
border={theme.borders.base}
px={2}
borderRadius={'md'}
>
<Avatar src={item.avatar} w={'24px'}></Avatar>
<Box ml={3} fontWeight={'bold'} fontSize={['md', 'lg', 'xl']}>
{item.name}
</Box>
</Flex>
))}
</Grid>
{isOpenKbSelect && (
<KBSelectModal
kbList={myKbList}
activeKbs={activeKbs}
onChange={onChange}
onClose={onCloseKbSelect}
/>
)}
</>
);
};
const NodeKbSearch = ({ data }: NodeProps<FlowModuleItemType>) => {
const { moduleId, inputs, outputs, onChangeNode } = data;
return (
<NodeCard minW={'400px'} {...data}>
<Divider text="Input" />
<Container>
<RenderInput
moduleId={moduleId}
onChangeNode={onChangeNode}
flowInputList={inputs}
CustomComponent={{
kbList: ({ key, value, ...props }) => (
<KBSelect
activeKbs={value}
onChange={(e) => {
onChangeNode({
moduleId,
key,
type: 'inputs',
value: {
...props,
key,
value: e
}
});
}}
/>
)
}}
/>
</Container>
<Divider text="Output" />
<Container>
<RenderOutput onChangeNode={onChangeNode} moduleId={moduleId} flowOutputList={outputs} />
</Container>
</NodeCard>
);
};
export default React.memo(NodeKbSearch);

View File

@@ -1,26 +0,0 @@
import React from 'react';
import { NodeProps } from 'reactflow';
import { Box } from '@chakra-ui/react';
import NodeCard from '../modules/NodeCard';
import { FlowModuleItemType } from '@/types/flow';
import Container from '../modules/Container';
import { SystemInputEnum } from '@/constants/app';
import { FlowValueTypeEnum } from '@/constants/flow';
import SourceHandle from '../render/SourceHandle';
const QuestionInputNode = ({ data }: NodeProps<FlowModuleItemType>) => {
return (
<NodeCard minW={'240px'} {...data}>
<Container borderTop={'2px solid'} borderTopColor={'myGray.200'} textAlign={'end'}>
<Box position={'relative'}>
<SourceHandle
handleKey={SystemInputEnum.userChatInput}
valueType={FlowValueTypeEnum.string}
/>
</Box>
</Container>
</NodeCard>
);
};
export default React.memo(QuestionInputNode);

View File

@@ -1,78 +0,0 @@
import React from 'react';
import { Handle, Position, NodeProps } from 'reactflow';
import { Flex, Box } from '@chakra-ui/react';
import NodeCard from '../modules/NodeCard';
import { SystemInputEnum } from '@/constants/app';
import { FlowModuleItemType } from '@/types/flow';
import Divider from '../modules/Divider';
import Container from '../modules/Container';
import Label from '../modules/Label';
const NodeTFSwitch = ({ data }: NodeProps<FlowModuleItemType>) => {
return (
<NodeCard minW={'220px'} {...data}>
<Divider text="输入输出" />
<Container h={'100px'} py={0} px={0} display={'flex'} alignItems={'center'}>
<Box flex={1} pl={'12px'}>
<Label
required
description="接收到 false、0、null、undefined或空字符串时执行 False反之执行 True"
>
</Label>
<Handle
style={{
top: '50%',
left: '0',
transform: 'translate(-50%,-50%)',
width: '12px',
height: '12px',
background: '#9CA2A8'
}}
id={SystemInputEnum.switch}
type="target"
position={Position.Left}
onConnect={(params) => console.log('input onConnect', params)}
/>
</Box>
<Box flex={1} pr={'12px'}>
<Flex alignItems={'center'} justifyContent={'flex-end'} mb={'26px'} position={'relative'}>
<Label>True</Label>
<Handle
style={{
top: '0',
right: '-12px',
transform: 'translate(50%,5px)',
width: '12px',
height: '12px',
background: '#9CA2A8'
}}
id={'true'}
type="source"
position={Position.Right}
onConnect={(params) => console.log('handle onConnect', params)}
/>
</Flex>
<Flex alignItems={'center'} justifyContent={'flex-end'} position={'relative'}>
<Label>False</Label>
<Handle
style={{
bottom: '0',
right: '-12px',
transform: 'translate(50%,-5px)',
width: '12px',
height: '12px',
background: '#9CA2A8'
}}
id={'false'}
type="source"
position={Position.Right}
onConnect={(params) => console.log('handle onConnect', params)}
/>
</Flex>
</Box>
</Container>
</NodeCard>
);
};
export default React.memo(NodeTFSwitch);

View File

@@ -1,59 +0,0 @@
import React, { useMemo } from 'react';
import { NodeProps } from 'reactflow';
import { Box, Flex, Textarea } from '@chakra-ui/react';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
import NodeCard from '../modules/NodeCard';
import { FlowModuleItemType } from '@/types/flow';
import Container from '../modules/Container';
import { SystemInputEnum } from '@/constants/app';
import MyIcon from '@/components/Icon';
import MyTooltip from '@/components/MyTooltip';
import { welcomeTextTip } from '@/constants/flow/ModuleTemplate';
const NodeUserGuide = ({ data }: NodeProps<FlowModuleItemType>) => {
const { inputs, moduleId, onChangeNode } = data;
const welcomeText = useMemo(
() => inputs.find((item) => item.key === SystemInputEnum.welcomeText),
[inputs]
);
return (
<>
<NodeCard minW={'300px'} {...data}>
<Container borderTop={'2px solid'} borderTopColor={'myGray.200'}>
<>
<Flex mb={1} alignItems={'center'}>
<MyIcon name={'welcomeText'} mr={2} w={'16px'} color={'#E74694'} />
<Box></Box>
<MyTooltip label={welcomeTextTip} forceShow>
<QuestionOutlineIcon display={['none', 'inline']} ml={1} />
</MyTooltip>
</Flex>
{welcomeText && (
<Textarea
className="nodrag"
rows={6}
resize={'both'}
defaultValue={welcomeText.value}
bg={'myWhite.500'}
placeholder={welcomeTextTip}
onChange={(e) => {
onChangeNode({
moduleId,
key: SystemInputEnum.welcomeText,
type: 'inputs',
value: {
...welcomeText,
value: e.target.value
}
});
}}
/>
)}
</>
</Container>
</NodeCard>
</>
);
};
export default React.memo(NodeUserGuide);

View File

@@ -1,126 +0,0 @@
import React, { useMemo } from 'react';
import { Box, Flex } from '@chakra-ui/react';
import { ModuleTemplates } from '@/constants/flow/ModuleTemplate';
import { FlowModuleItemType, FlowModuleTemplateType } from '@/types/flow';
import type { Node, XYPosition } from 'reactflow';
import { useGlobalStore } from '@/store/global';
import Avatar from '@/components/Avatar';
import { FlowModuleTypeEnum } from '@/constants/flow';
const ModuleTemplateList = ({
nodes,
isOpen,
onAddNode,
onClose
}: {
nodes?: Node<FlowModuleItemType>[];
isOpen: boolean;
onAddNode: (e: { template: FlowModuleTemplateType; position: XYPosition }) => void;
onClose: () => void;
}) => {
const { isPc } = useGlobalStore();
const filterTemplates = useMemo(() => {
const guideModulesIndex = ModuleTemplates.findIndex((item) => item.label === '引导模块');
const guideModule: {
label: string;
list: FlowModuleTemplateType[];
} = JSON.parse(JSON.stringify(ModuleTemplates[guideModulesIndex]));
if (nodes?.find((item) => item.type === FlowModuleTypeEnum.userGuide)) {
const index = guideModule.list.findIndex(
(item) => item.flowType === FlowModuleTypeEnum.userGuide
);
guideModule.list.splice(index, 1);
}
if (nodes?.find((item) => item.type === FlowModuleTypeEnum.variable)) {
const index = guideModule.list.findIndex(
(item) => item.flowType === FlowModuleTypeEnum.variable
);
guideModule.list.splice(index, 1);
}
return [
...ModuleTemplates.slice(0, guideModulesIndex),
guideModule,
...ModuleTemplates.slice(guideModulesIndex + 1)
];
}, [nodes]);
return (
<>
<Box
zIndex={2}
display={isOpen ? 'block' : 'none'}
position={'absolute'}
top={0}
left={0}
bottom={0}
w={'360px'}
onClick={onClose}
/>
<Flex
zIndex={3}
flexDirection={'column'}
position={'absolute'}
top={'65px'}
left={0}
pb={4}
h={isOpen ? 'calc(100% - 100px)' : '0'}
w={isOpen ? ['100%', '360px'] : '0'}
bg={'white'}
boxShadow={'3px 0 20px rgba(0,0,0,0.2)'}
borderRadius={'20px'}
overflow={'hidden'}
transition={'.2s ease'}
userSelect={'none'}
>
<Box w={['100%', '330px']} py={4} px={5} fontSize={'xl'} fontWeight={'bold'}>
</Box>
<Box flex={'1 0 0'} overflow={'overlay'}>
<Box w={['100%', '330px']} mx={'auto'}>
{filterTemplates.map((item) =>
item.list.map((item) => (
<Flex
key={item.name}
alignItems={'center'}
p={5}
cursor={'pointer'}
_hover={{ bg: 'myWhite.600' }}
borderRadius={'md'}
draggable
onDragEnd={(e) => {
if (e.clientX < 360) return;
onAddNode({
template: item,
position: { x: e.clientX, y: e.clientY }
});
}}
onClick={(e) => {
if (isPc) return;
onClose();
onAddNode({
template: item,
position: { x: e.clientX, y: e.clientY }
});
}}
>
<Avatar src={item.logo} w={'34px'} objectFit={'contain'} borderRadius={'0'} />
<Box ml={5} flex={'1 0 0'}>
<Box color={'black'}>{item.name}</Box>
<Box color={'myGray.500'} fontSize={'sm'}>
{item.intro}
</Box>
</Box>
</Flex>
))
)}
</Box>
</Box>
</Flex>
</>
);
};
export default ModuleTemplateList;

View File

@@ -1,30 +0,0 @@
import React from 'react';
import { Box } from '@chakra-ui/react';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
import MyTooltip from '@/components/MyTooltip';
const Label = ({
required = false,
children,
description
}: {
required?: boolean;
children: React.ReactNode | string;
description?: string;
}) => (
<Box as={'label'} display={'inline-block'} position={'relative'}>
{children}
{required && (
<Box position={'absolute'} top={'-2px'} right={'-10px'} color={'red.500'} fontWeight={'bold'}>
*
</Box>
)}
{description && (
<MyTooltip label={description} forceShow>
<QuestionOutlineIcon display={['none', 'inline']} fontSize={'12px'} mb={1} ml={1} />
</MyTooltip>
)}
</Box>
);
export default React.memo(Label);

View File

@@ -1,107 +0,0 @@
import React, { useMemo } from 'react';
import { Box, Flex, useTheme, Menu, MenuButton, MenuList, MenuItem } from '@chakra-ui/react';
import MyIcon from '@/components/Icon';
import Avatar from '@/components/Avatar';
import type { FlowModuleItemType } from '@/types/flow';
import MyTooltip from '@/components/MyTooltip';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
import { useTranslation } from 'react-i18next';
import { useCopyData } from '@/utils/tools';
type Props = FlowModuleItemType & {
children?: React.ReactNode | React.ReactNode[] | string;
minW?: string | number;
};
const NodeCard = (props: Props) => {
const {
children,
logo = '/icon/logo.svg',
name = '未知模块',
description,
minW = '300px',
onCopyNode,
onDelNode,
moduleId
} = props;
const { copyData } = useCopyData();
const { t } = useTranslation();
const theme = useTheme();
const menuList = useMemo(
() => [
{
icon: 'copy',
label: t('common.Copy'),
onClick: () => onCopyNode(moduleId)
},
// {
// icon: 'settingLight',
// label: t('app.Copy Module Config'),
// onClick: () => {
// const copyProps = { ...props };
// delete copyProps.children;
// delete copyProps.children;
// console.log(copyProps);
// }
// },
{
icon: 'delete',
label: t('common.Delete'),
onClick: () => onDelNode(moduleId)
},
{
icon: 'back',
label: t('common.Cancel'),
onClick: () => {}
}
],
[moduleId, onCopyNode, onDelNode, t]
);
return (
<Box minW={minW} bg={'white'} border={theme.borders.md} borderRadius={'md'} boxShadow={'sm'}>
<Flex className="custom-drag-handle" px={4} py={3} alignItems={'center'}>
<Avatar src={logo} borderRadius={'md'} objectFit={'contain'} w={'30px'} h={'30px'} />
<Box ml={3} fontSize={'lg'} color={'myGray.600'}>
{name}
</Box>
{description && (
<MyTooltip label={description} forceShow>
<QuestionOutlineIcon
display={['none', 'inline']}
transform={'translateY(1px)'}
ml={1}
/>
</MyTooltip>
)}
<Box flex={1} />
<Menu autoSelect={false} isLazy>
<MenuButton
className={'nodrag'}
_hover={{ bg: 'myWhite.600' }}
cursor={'pointer'}
borderRadius={'md'}
onClick={(e) => {
e.stopPropagation();
}}
>
<MyIcon name={'more'} w={'14px'} p={2} />
</MenuButton>
<MenuList color={'myGray.700'} minW={`120px !important`} zIndex={10}>
{menuList.map((item) => (
<MenuItem key={item.label} onClick={item.onClick} py={[2, 3]}>
<MyIcon name={item.icon as any} w={['14px', '16px']} />
<Box ml={[1, 2]}>{item.label}</Box>
</MenuItem>
))}
</MenuList>
</Menu>
</Flex>
{children}
</Box>
);
};
export default React.memo(NodeCard);

View File

@@ -1,115 +0,0 @@
import React, { useMemo, useState } from 'react';
import {
Box,
Button,
ModalHeader,
ModalFooter,
ModalBody,
Flex,
Switch,
Input,
FormControl
} from '@chakra-ui/react';
import { useForm } from 'react-hook-form';
import { customAlphabet } from 'nanoid';
const nanoid = customAlphabet('abcdefghijklmnopqrstuvwxyz1234567890', 6);
import MyModal from '@/components/MyModal';
import Avatar from '@/components/Avatar';
import MyTooltip from '@/components/MyTooltip';
import { FlowInputItemTypeEnum, FlowValueTypeEnum } from '@/constants/flow';
import { useTranslation } from 'react-i18next';
import MySelect from '@/components/Select';
import { FlowInputItemType } from '@/types/flow';
const typeSelectList = [
{
label: '字符串',
value: FlowValueTypeEnum.string
},
{
label: '数字',
value: FlowValueTypeEnum.number
},
{
label: '布尔',
value: FlowValueTypeEnum.boolean
},
{
label: '任意',
value: FlowValueTypeEnum.any
}
];
const SetInputFieldModal = ({
defaultField = {
label: '',
key: '',
type: FlowInputItemTypeEnum.target,
valueType: FlowValueTypeEnum.string,
description: '',
required: false
},
onClose,
onSubmit
}: {
defaultField?: FlowInputItemType;
onClose: () => void;
onSubmit: (data: FlowInputItemType) => void;
}) => {
const { t } = useTranslation();
const { register, getValues, setValue, handleSubmit } = useForm<FlowInputItemType>({
defaultValues: defaultField
});
const [refresh, setRefresh] = useState(false);
return (
<MyModal isOpen={true} onClose={onClose}>
<ModalHeader display={'flex'} alignItems={'center'}>
<Avatar src={'/imgs/module/extract.png'} mr={2} w={'20px'} objectFit={'cover'} />
{t('app.Input Field Settings')}
</ModalHeader>
<ModalBody>
<Flex alignItems={'center'}>
<Box flex={'0 0 70px'}></Box>
<Switch {...register('required')} />
</Flex>
<Flex mt={5} alignItems={'center'}>
<Box flex={'0 0 70px'}></Box>
<MySelect
w={'288px'}
list={typeSelectList}
value={getValues('valueType')}
onchange={(e: any) => {
setValue('valueType', e);
setRefresh(!refresh);
}}
/>
</Flex>
<Flex mt={5} alignItems={'center'}>
<Box flex={'0 0 70px'}></Box>
<Input
placeholder="预约字段/sql语句……"
{...register('label', { required: '字段名不能为空' })}
/>
</Flex>
<Flex mt={5} alignItems={'center'}>
<Box flex={'0 0 70px'}> key</Box>
<Input
placeholder="appointment/sql"
{...register('key', { required: '字段 key 不能为空' })}
/>
</Flex>
</ModalBody>
<ModalFooter>
<Button variant={'base'} mr={3} onClick={onClose}>
</Button>
<Button onClick={handleSubmit(onSubmit)}></Button>
</ModalFooter>
</MyModal>
);
};
export default React.memo(SetInputFieldModal);

View File

@@ -1,105 +0,0 @@
import React, { useMemo, useState } from 'react';
import {
Box,
Button,
ModalHeader,
ModalFooter,
ModalBody,
Flex,
Switch,
Input,
FormControl
} from '@chakra-ui/react';
import type { ContextExtractAgentItemType, HttpFieldItemType } from '@/types/app';
import { useForm } from 'react-hook-form';
import { customAlphabet } from 'nanoid';
const nanoid = customAlphabet('abcdefghijklmnopqrstuvwxyz1234567890', 6);
import MyModal from '@/components/MyModal';
import Avatar from '@/components/Avatar';
import MyTooltip from '@/components/MyTooltip';
import { FlowOutputItemTypeEnum, FlowValueTypeEnum, FlowValueTypeStyle } from '@/constants/flow';
import { useTranslation } from 'react-i18next';
import MySelect from '@/components/Select';
import { FlowOutputItemType } from '@/types/flow';
const typeSelectList = [
{
label: '字符串',
value: FlowValueTypeEnum.string
},
{
label: '数字',
value: FlowValueTypeEnum.number
},
{
label: '布尔',
value: FlowValueTypeEnum.boolean
},
{
label: '任意',
value: FlowValueTypeEnum.any
}
];
const SetInputFieldModal = ({
defaultField,
onClose,
onSubmit
}: {
defaultField: FlowOutputItemType;
onClose: () => void;
onSubmit: (data: FlowOutputItemType) => void;
}) => {
const { t } = useTranslation();
const { register, getValues, setValue, handleSubmit } = useForm<FlowOutputItemType>({
defaultValues: defaultField
});
const [refresh, setRefresh] = useState(false);
return (
<MyModal isOpen={true} onClose={onClose}>
<ModalHeader display={'flex'} alignItems={'center'}>
<Avatar src={'/imgs/module/extract.png'} mr={2} w={'20px'} objectFit={'cover'} />
{t('app.Output Field Settings')}
</ModalHeader>
<ModalBody>
<Flex mt={5} alignItems={'center'}>
<Box flex={'0 0 70px'}></Box>
<MySelect
w={'288px'}
list={typeSelectList}
value={getValues('valueType')}
onchange={(e: any) => {
setValue('valueType', e);
setRefresh(!refresh);
}}
/>
</Flex>
<Flex mt={5} alignItems={'center'}>
<Box flex={'0 0 70px'}></Box>
<Input
placeholder="预约字段/sql语句……"
{...register('label', { required: '字段名不能为空' })}
/>
</Flex>
<Flex mt={5} alignItems={'center'}>
<Box flex={'0 0 70px'}> key</Box>
<Input
placeholder="appointment/sql"
{...register('key', { required: '字段 key 不能为空' })}
/>
</Flex>
</ModalBody>
<ModalFooter>
<Button variant={'base'} mr={3} onClick={onClose}>
</Button>
<Button onClick={handleSubmit(onSubmit)}></Button>
</ModalFooter>
</MyModal>
);
};
export default React.memo(SetInputFieldModal);

View File

@@ -1,272 +0,0 @@
import React, { useState } from 'react';
import type { FlowInputItemType, FlowModuleItemType } from '@/types/flow';
import {
Box,
Textarea,
Input,
NumberInput,
NumberInputField,
NumberInputStepper,
NumberIncrementStepper,
NumberDecrementStepper,
Flex
} from '@chakra-ui/react';
import { FlowInputItemTypeEnum } from '@/constants/flow';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
import dynamic from 'next/dynamic';
import MySelect from '@/components/Select';
import MySlider from '@/components/Slider';
import MyTooltip from '@/components/MyTooltip';
import TargetHandle from './TargetHandle';
import MyIcon from '@/components/Icon';
const SetInputFieldModal = dynamic(() => import('../modules/SetInputFieldModal'));
export const Label = ({
moduleId,
inputKey,
onChangeNode,
...item
}: FlowInputItemType & {
moduleId: string;
inputKey: string;
onChangeNode: FlowModuleItemType['onChangeNode'];
}) => {
const { required = false, description, edit, label, type, valueType } = item;
const [editField, setEditField] = useState<FlowInputItemType>();
return (
<Flex className="nodrag" cursor={'default'} alignItems={'center'} position={'relative'}>
<Box position={'relative'}>
{label}
{description && (
<MyTooltip label={description} forceShow>
<QuestionOutlineIcon display={['none', 'inline']} ml={1} />
</MyTooltip>
)}
{required && (
<Box
position={'absolute'}
top={'-2px'}
right={'-8px'}
color={'red.500'}
fontWeight={'bold'}
>
*
</Box>
)}
</Box>
{(type === FlowInputItemTypeEnum.target || valueType) && (
<TargetHandle handleKey={inputKey} valueType={valueType} />
)}
{edit && (
<>
<MyIcon
name={'settingLight'}
w={'14px'}
cursor={'pointer'}
ml={3}
_hover={{ color: 'myBlue.600' }}
onClick={() =>
setEditField({
...item,
key: inputKey
})
}
/>
<MyIcon
className="delete"
name={'delete'}
w={'14px'}
cursor={'pointer'}
ml={2}
_hover={{ color: 'red.500' }}
onClick={() => {
onChangeNode({
moduleId,
type: 'delInput',
key: inputKey,
value: ''
});
}}
/>
</>
)}
{!!editField && (
<SetInputFieldModal
defaultField={editField}
onClose={() => setEditField(undefined)}
onSubmit={(data) => {
// same key
if (editField.key === data.key) {
onChangeNode({
moduleId,
type: 'inputs',
key: inputKey,
value: data
});
} else {
// diff key. del and add
onChangeNode({
moduleId,
type: 'addInput',
key: data.key,
value: data
});
setTimeout(() => {
onChangeNode({
moduleId,
type: 'delInput',
key: editField.key,
value: ''
});
});
}
setEditField(undefined);
}}
/>
)}
</Flex>
);
};
const RenderInput = ({
flowInputList,
moduleId,
CustomComponent = {},
onChangeNode
}: {
flowInputList: FlowInputItemType[];
moduleId: string;
CustomComponent?: Record<string, (e: FlowInputItemType) => React.ReactNode>;
onChangeNode: FlowModuleItemType['onChangeNode'];
}) => {
return (
<>
{flowInputList.map(
(item) =>
item.type !== FlowInputItemTypeEnum.hidden && (
<Box key={item.key} _notLast={{ mb: 7 }} position={'relative'}>
{!!item.label && (
<Label
moduleId={moduleId}
onChangeNode={onChangeNode}
inputKey={item.key}
{...item}
/>
)}
<Box mt={2} className={'nodrag'}>
{item.type === FlowInputItemTypeEnum.numberInput && (
<NumberInput
defaultValue={item.value}
min={item.min}
max={item.max}
onChange={(e) => {
onChangeNode({
moduleId,
type: 'inputs',
key: item.key,
value: {
...item,
value: Number(e)
}
});
}}
>
<NumberInputField />
<NumberInputStepper>
<NumberIncrementStepper />
<NumberDecrementStepper />
</NumberInputStepper>
</NumberInput>
)}
{item.type === FlowInputItemTypeEnum.input && (
<Input
placeholder={item.placeholder}
defaultValue={item.value}
onChange={(e) => {
onChangeNode({
moduleId,
type: 'inputs',
key: item.key,
value: {
...item,
value: e.target.value
}
});
}}
/>
)}
{item.type === FlowInputItemTypeEnum.textarea && (
<Textarea
rows={5}
placeholder={item.placeholder}
resize={'both'}
defaultValue={item.value}
onChange={(e) => {
onChangeNode({
moduleId,
type: 'inputs',
key: item.key,
value: {
...item,
value: e.target.value
}
});
}}
/>
)}
{item.type === FlowInputItemTypeEnum.select && (
<MySelect
width={'100%'}
value={item.value}
list={item.list || []}
onchange={(e) => {
onChangeNode({
moduleId,
type: 'inputs',
key: item.key,
value: {
...item,
value: e
}
});
}}
/>
)}
{item.type === FlowInputItemTypeEnum.slider && (
<Box pt={5} pb={4} px={2}>
<MySlider
markList={item.markList}
width={'100%'}
min={item.min || 0}
max={item.max}
step={item.step || 1}
value={item.value}
onChange={(e) => {
onChangeNode({
moduleId,
type: 'inputs',
key: item.key,
value: {
...item,
value: e
}
});
}}
/>
</Box>
)}
{item.type === FlowInputItemTypeEnum.custom && CustomComponent[item.key] && (
<>{CustomComponent[item.key]({ ...item })}</>
)}
</Box>
</Box>
)
)}
</>
);
};
export default React.memo(RenderInput);

View File

@@ -1,5 +0,0 @@
.panel {
.react-flow__panel {
display: none;
}
}

View File

@@ -1,632 +0,0 @@
import React, { useCallback, useEffect, useRef, useState } from 'react';
import ReactFlow, {
Background,
Controls,
ReactFlowProvider,
addEdge,
useNodesState,
useEdgesState,
XYPosition,
Connection,
useViewport
} from 'reactflow';
import { Box, Flex, IconButton, useTheme, useDisclosure } from '@chakra-ui/react';
import { SmallCloseIcon } from '@chakra-ui/icons';
import {
edgeOptions,
connectionLineStyle,
FlowModuleTypeEnum,
FlowInputItemTypeEnum,
FlowValueTypeEnum
} from '@/constants/flow';
import { appModule2FlowNode, appModule2FlowEdge } from '@/utils/adapt';
import {
FlowModuleItemType,
FlowModuleTemplateType,
FlowOutputTargetItemType,
type FlowModuleItemChangeProps
} from '@/types/flow';
import { AppModuleItemType } from '@/types/app';
import { customAlphabet } from 'nanoid';
import { useRequest } from '@/hooks/useRequest';
import type { AppSchema } from '@/types/mongoSchema';
import { useUserStore } from '@/store/user';
import { useToast } from '@/hooks/useToast';
import { useTranslation } from 'next-i18next';
import { useCopyData } from '@/utils/tools';
import dynamic from 'next/dynamic';
import MyIcon from '@/components/Icon';
import ButtonEdge from './components/modules/ButtonEdge';
import MyTooltip from '@/components/MyTooltip';
import TemplateList from './components/TemplateList';
import ChatTest, { type ChatTestComponentRef } from './components/ChatTest';
const ImportSettings = dynamic(() => import('./components/ImportSettings'), {
ssr: false
});
const NodeChat = dynamic(() => import('./components/Nodes/NodeChat'), {
ssr: false
});
const NodeKbSearch = dynamic(() => import('./components/Nodes/NodeKbSearch'), {
ssr: false
});
const NodeHistory = dynamic(() => import('./components/Nodes/NodeHistory'), {
ssr: false
});
const NodeTFSwitch = dynamic(() => import('./components/Nodes/NodeTFSwitch'), {
ssr: false
});
const NodeAnswer = dynamic(() => import('./components/Nodes/NodeAnswer'), {
ssr: false
});
const NodeQuestionInput = dynamic(() => import('./components/Nodes/NodeQuestionInput'), {
ssr: false
});
const NodeCQNode = dynamic(() => import('./components/Nodes/NodeCQNode'), {
ssr: false
});
const NodeVariable = dynamic(() => import('./components/Nodes/NodeVariable'), {
ssr: false
});
const NodeUserGuide = dynamic(() => import('./components/Nodes/NodeUserGuide'), {
ssr: false
});
const NodeExtract = dynamic(() => import('./components/Nodes/NodeExtract'), {
ssr: false
});
const NodeHttp = dynamic(() => import('./components/Nodes/NodeHttp'), {
ssr: false
});
import 'reactflow/dist/style.css';
import styles from './index.module.scss';
import { AppTypeEnum } from '@/constants/app';
const nanoid = customAlphabet('abcdefghijklmnopqrstuvwxyz1234567890', 6);
const nodeTypes = {
[FlowModuleTypeEnum.userGuide]: NodeUserGuide,
[FlowModuleTypeEnum.variable]: NodeVariable,
[FlowModuleTypeEnum.questionInput]: NodeQuestionInput,
[FlowModuleTypeEnum.historyNode]: NodeHistory,
[FlowModuleTypeEnum.chatNode]: NodeChat,
[FlowModuleTypeEnum.kbSearchNode]: NodeKbSearch,
[FlowModuleTypeEnum.tfSwitchNode]: NodeTFSwitch,
[FlowModuleTypeEnum.answerNode]: NodeAnswer,
[FlowModuleTypeEnum.classifyQuestion]: NodeCQNode,
[FlowModuleTypeEnum.contentExtract]: NodeExtract,
[FlowModuleTypeEnum.httpRequest]: NodeHttp
// [FlowModuleTypeEnum.empty]: EmptyModule
};
const edgeTypes = {
buttonedge: ButtonEdge
};
type Props = { app: AppSchema; onCloseSettings: () => void };
const AppEdit = ({ app, onCloseSettings }: Props) => {
const theme = useTheme();
const { toast } = useToast();
const { t } = useTranslation();
const { copyData } = useCopyData();
const reactFlowWrapper = useRef<HTMLDivElement>(null);
const ChatTestRef = useRef<ChatTestComponentRef>(null);
const { updateAppDetail } = useUserStore();
const { x, y, zoom } = useViewport();
const [nodes, setNodes, onNodesChange] = useNodesState<FlowModuleItemType>([]);
const [edges, setEdges, onEdgesChange] = useEdgesState([]);
const {
isOpen: isOpenTemplate,
onOpen: onOpenTemplate,
onClose: onCloseTemplate
} = useDisclosure();
const { isOpen: isOpenImport, onOpen: onOpenImport, onClose: onCloseImport } = useDisclosure();
const [testModules, setTestModules] = useState<AppModuleItemType[]>();
const onFixView = useCallback(() => {
const btn = document.querySelector('.react-flow__controls-fitview') as HTMLButtonElement;
setTimeout(() => {
btn && btn.click();
}, 100);
}, []);
const onAddNode = useCallback(
({ template, position }: { template: FlowModuleTemplateType; position: XYPosition }) => {
if (!reactFlowWrapper.current) return;
const reactFlowBounds = reactFlowWrapper.current.getBoundingClientRect();
const mouseX = (position.x - reactFlowBounds.left - x) / zoom - 100;
const mouseY = (position.y - reactFlowBounds.top - y) / zoom;
setNodes((state) =>
state.concat(
appModule2FlowNode({
item: {
...template,
moduleId: nanoid(),
position: { x: mouseX, y: mouseY }
},
onChangeNode,
onDelNode,
onDelEdge,
onCopyNode,
onCollectionNode
})
)
);
},
[x, zoom, y]
);
const onDelNode = useCallback(
(nodeId: string) => {
setNodes((state) => state.filter((item) => item.id !== nodeId));
setEdges((state) => state.filter((edge) => edge.source !== nodeId && edge.target !== nodeId));
},
[setEdges, setNodes]
);
const onDelEdge = useCallback(
({
moduleId,
sourceHandle,
targetHandle
}: {
moduleId: string;
sourceHandle?: string;
targetHandle?: string;
}) => {
if (!sourceHandle && !targetHandle) return;
setEdges((state) =>
state.filter((edge) => {
if (edge.source === moduleId && edge.sourceHandle === sourceHandle) return false;
if (edge.target === moduleId && edge.targetHandle === targetHandle) return false;
return true;
})
);
},
[setEdges]
);
const onCopyNode = useCallback(
(nodeId: string) => {
setNodes((nodes) => {
const node = nodes.find((node) => node.id === nodeId);
if (!node) return nodes;
const template = {
logo: node.data.logo,
name: node.data.name,
intro: node.data.intro,
description: node.data.description,
flowType: node.data.flowType,
inputs: node.data.inputs,
outputs: node.data.outputs,
showStatus: node.data.showStatus
};
return nodes.concat(
appModule2FlowNode({
item: {
...template,
moduleId: nanoid(),
position: { x: node.position.x + 200, y: node.position.y + 50 }
},
onChangeNode,
onDelNode,
onDelEdge,
onCopyNode,
onCollectionNode
})
);
});
},
[setNodes]
);
const onCollectionNode = useCallback(
(nodeId: string) => {
console.log(nodes.find((node) => node.id === nodeId));
},
[nodes]
);
const flow2AppModules = useCallback(() => {
const modules: AppModuleItemType[] = nodes.map((item) => ({
moduleId: item.data.moduleId,
name: item.data.name,
flowType: item.data.flowType,
showStatus: item.data.showStatus,
position: item.position,
inputs: item.data.inputs.map((item) => ({
...item,
connected: item.type !== FlowInputItemTypeEnum.target
})),
outputs: item.data.outputs.map((item) => ({
...item,
targets: [] as FlowOutputTargetItemType[]
}))
}));
// update inputs and outputs
modules.forEach((module) => {
module.inputs.forEach((input) => {
input.connected =
input.connected ||
!!edges.find(
(edge) => edge.target === module.moduleId && edge.targetHandle === input.key
);
});
module.outputs.forEach((output) => {
output.targets = edges
.filter(
(edge) =>
edge.source === module.moduleId &&
edge.sourceHandle === output.key &&
edge.targetHandle
)
.map((edge) => ({
moduleId: edge.target,
key: edge.targetHandle || ''
}));
});
});
return modules;
}, [edges, nodes]);
const onChangeNode = useCallback(
({ moduleId, key, type = 'inputs', value }: FlowModuleItemChangeProps) => {
setNodes((nodes) =>
nodes.map((node) => {
if (node.id !== moduleId) return node;
if (type === 'inputs') {
return {
...node,
data: {
...node.data,
inputs: node.data.inputs.map((item) => (item.key === key ? value : item))
}
};
}
if (type === 'addInput') {
const input = node.data.inputs.find((input) => input.key === value.key);
if (input) {
toast({
status: 'warning',
title: 'key 重复'
});
return {
...node,
data: {
...node.data,
inputs: node.data.inputs
}
};
}
return {
...node,
data: {
...node.data,
inputs: node.data.inputs.concat(value)
}
};
}
if (type === 'delInput') {
onDelEdge({ moduleId, targetHandle: key });
return {
...node,
data: {
...node.data,
inputs: node.data.inputs.filter((item) => item.key !== key)
}
};
}
// del output connect
const delOutputs = node.data.outputs.filter(
(item) => !value.find((output: FlowOutputTargetItemType) => output.key === item.key)
);
delOutputs.forEach((output) => {
onDelEdge({ moduleId, sourceHandle: output.key });
});
return {
...node,
data: {
...node.data,
outputs: value
}
};
})
);
},
[]
);
const onDelConnect = useCallback((id: string) => {
setEdges((state) => state.filter((item) => item.id !== id));
}, []);
const onConnect = useCallback(
({ connect }: { connect: Connection }) => {
const source = nodes.find((node) => node.id === connect.source)?.data;
const sourceType = (() => {
if (source?.flowType === FlowModuleTypeEnum.classifyQuestion) {
return FlowValueTypeEnum.boolean;
}
return source?.outputs.find((output) => output.key === connect.sourceHandle)?.valueType;
})();
const targetType = nodes
.find((node) => node.id === connect.target)
?.data?.inputs.find((input) => input.key === connect.targetHandle)?.valueType;
if (!sourceType || !targetType) {
return toast({
status: 'warning',
title: t('app.Connection is invalid')
});
}
if (
sourceType !== FlowValueTypeEnum.any &&
targetType !== FlowValueTypeEnum.any &&
sourceType !== targetType
) {
return toast({
status: 'warning',
title: t('app.Connection type is different')
});
}
setEdges((state) =>
addEdge(
{
...connect,
type: 'buttonedge',
animated: true,
data: {
onDelete: onDelConnect
}
},
state
)
);
},
[nodes]
);
const { mutate: onclickSave, isLoading } = useRequest({
mutationFn: () => {
return updateAppDetail(app._id, {
modules: flow2AppModules(),
type: AppTypeEnum.advanced
});
},
successToast: '保存配置成功',
errorToast: '保存配置异常',
onSuccess() {
ChatTestRef.current?.resetChatTest();
}
});
const initData = useCallback(
(modules: AppModuleItemType[]) => {
const edges = appModule2FlowEdge({
modules,
onDelete: onDelConnect
});
setEdges(edges);
setNodes(
modules.map((item) =>
appModule2FlowNode({
item,
onChangeNode,
onDelNode,
onDelEdge,
onCopyNode,
onCollectionNode
})
)
);
onFixView();
},
[
onDelConnect,
setEdges,
setNodes,
onFixView,
onChangeNode,
onDelNode,
onDelEdge,
onCopyNode,
onCollectionNode
]
);
useEffect(() => {
initData(JSON.parse(JSON.stringify(app.modules)));
}, [app.modules]);
return (
<>
{/* header */}
<Flex
py={3}
px={[2, 5, 8]}
borderBottom={theme.borders.base}
alignItems={'center'}
userSelect={'none'}
>
<MyTooltip label={'返回'} offset={[10, 10]}>
<IconButton
size={'sm'}
icon={<MyIcon name={'back'} w={'14px'} />}
borderRadius={'md'}
borderColor={'myGray.300'}
variant={'base'}
aria-label={''}
onClick={() => {
onCloseSettings();
onFixView();
}}
/>
</MyTooltip>
<Box ml={[3, 6]} fontSize={['md', '2xl']} flex={1}>
{app.name}
</Box>
<MyTooltip label={t('app.Import Configs')}>
<IconButton
mr={[3, 6]}
icon={<MyIcon name={'importLight'} w={['14px', '16px']} />}
borderRadius={'lg'}
variant={'base'}
aria-label={'save'}
onClick={onOpenImport}
/>
</MyTooltip>
<MyTooltip label={t('app.Export Configs')}>
<IconButton
mr={[3, 6]}
icon={<MyIcon name={'export'} w={['14px', '16px']} />}
borderRadius={'lg'}
variant={'base'}
aria-label={'save'}
onClick={() =>
copyData(
JSON.stringify(flow2AppModules(), null, 2),
t('app.Export Config Successful')
)
}
/>
</MyTooltip>
{testModules ? (
<IconButton
mr={[3, 6]}
icon={<SmallCloseIcon fontSize={'25px'} />}
variant={'base'}
color={'myGray.600'}
borderRadius={'lg'}
aria-label={''}
onClick={() => setTestModules(undefined)}
/>
) : (
<MyTooltip label={'测试对话'}>
<IconButton
mr={[3, 6]}
icon={<MyIcon name={'chat'} w={['14px', '16px']} />}
borderRadius={'lg'}
aria-label={'save'}
variant={'base'}
onClick={() => {
setTestModules(flow2AppModules());
}}
/>
</MyTooltip>
)}
<MyTooltip label={'保存配置'}>
<IconButton
icon={<MyIcon name={'save'} w={['14px', '16px']} />}
borderRadius={'lg'}
isLoading={isLoading}
aria-label={'save'}
onClick={onclickSave}
/>
</MyTooltip>
</Flex>
<Box
minH={'400px'}
flex={'1 0 0'}
w={'100%'}
h={0}
position={'relative'}
onContextMenu={(e) => {
e.preventDefault();
return false;
}}
>
{/* open module template */}
<IconButton
position={'absolute'}
top={5}
left={5}
w={'38px'}
h={'38px'}
borderRadius={'50%'}
icon={<SmallCloseIcon fontSize={'26px'} />}
transform={isOpenTemplate ? '' : 'rotate(135deg)'}
transition={'0.2s ease'}
aria-label={''}
zIndex={1}
boxShadow={'2px 2px 6px #85b1ff'}
onClick={() => {
isOpenTemplate ? onCloseTemplate() : onOpenTemplate();
}}
/>
<ReactFlow
ref={reactFlowWrapper}
className={styles.panel}
fitView
nodes={nodes}
edges={edges}
minZoom={0.4}
maxZoom={1.5}
defaultEdgeOptions={edgeOptions}
connectionLineStyle={connectionLineStyle}
nodeTypes={nodeTypes}
edgeTypes={edgeTypes}
onNodesChange={onNodesChange}
onEdgesChange={onEdgesChange}
onConnect={(connect) => {
connect.sourceHandle &&
connect.targetHandle &&
onConnect({
connect
});
}}
>
<Background />
<Controls position={'bottom-right'} style={{ display: 'flex' }} showInteractive={false} />
</ReactFlow>
<TemplateList
isOpen={isOpenTemplate}
nodes={nodes}
onAddNode={onAddNode}
onClose={onCloseTemplate}
/>
<ChatTest
ref={ChatTestRef}
modules={testModules}
app={app}
onClose={() => setTestModules(undefined)}
/>
</Box>
{isOpenImport && (
<ImportSettings
onClose={onCloseImport}
onSuccess={(data) => {
setEdges([]);
setNodes([]);
setTimeout(() => {
initData(data);
}, 10);
}}
/>
)}
</>
);
};
const Flow = (data: Props) => (
<Box h={'100%'} position={'fixed'} zIndex={999} top={0} left={0} right={0} bottom={0}>
<ReactFlowProvider>
<Flex h={'100%'} flexDirection={'column'} bg={'#fff'}>
{!!data.app._id && <AppEdit {...data} />}
</Flex>
</ReactFlowProvider>
</Box>
);
export default React.memo(Flow);

View File

@@ -1,242 +0,0 @@
import React, { useState } from 'react';
import {
Card,
Flex,
Box,
Button,
ModalBody,
ModalHeader,
ModalFooter,
useTheme,
Textarea
} from '@chakra-ui/react';
import Avatar from '@/components/Avatar';
import { KbListItemType } from '@/types/plugin';
import { useForm } from 'react-hook-form';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
import type { SelectedKbType } from '@/types/plugin';
import { useGlobalStore } from '@/store/global';
import { useToast } from '@/hooks/useToast';
import MySlider from '@/components/Slider';
import MyTooltip from '@/components/MyTooltip';
import MyModal from '@/components/MyModal';
import MyIcon from '@/components/Icon';
export type KbParamsType = {
searchSimilarity: number;
searchLimit: number;
searchEmptyText: string;
};
export const KBSelectModal = ({
kbList,
activeKbs = [],
onChange,
onClose
}: {
kbList: KbListItemType[];
activeKbs: SelectedKbType;
onChange: (e: SelectedKbType) => void;
onClose: () => void;
}) => {
const theme = useTheme();
const [selectedKbList, setSelectedKbList] = useState<SelectedKbType>(activeKbs);
const { isPc } = useGlobalStore();
const { toast } = useToast();
return (
<MyModal
isOpen={true}
isCentered={!isPc}
maxW={['90vw', '800px']}
w={'800px'}
onClose={onClose}
>
<Flex flexDirection={'column'} h={['90vh', 'auto']}>
<ModalHeader>
<Box>({selectedKbList.length})</Box>
<Box fontSize={'sm'} color={'myGray.500'} fontWeight={'normal'}>
</Box>
</ModalHeader>
<ModalBody
flex={['1 0 0', '0 0 auto']}
maxH={'80vh'}
overflowY={'auto'}
display={'grid'}
gridTemplateColumns={['repeat(1,1fr)', 'repeat(2,1fr)', 'repeat(3,1fr)']}
gridGap={3}
userSelect={'none'}
>
{kbList.map((item) =>
(() => {
const selected = !!selectedKbList.find((kb) => kb.kbId === item._id);
const active = !!activeKbs.find((kb) => kb.kbId === item._id);
return (
<Card
key={item._id}
p={3}
border={theme.borders.base}
boxShadow={'sm'}
h={'80px'}
cursor={'pointer'}
order={active ? 0 : 1}
_hover={{
boxShadow: 'md'
}}
{...(selected
? {
bg: 'myBlue.300'
}
: {})}
onClick={() => {
if (selected) {
setSelectedKbList((state) => state.filter((kb) => kb.kbId !== item._id));
} else {
const vectorModel = selectedKbList[0]?.vectorModel?.model;
if (vectorModel && vectorModel !== item.vectorModel.model) {
return toast({
status: 'warning',
title: '仅能选择同一个索引模型的知识库'
});
}
setSelectedKbList((state) => [
...state,
{ kbId: item._id, vectorModel: item.vectorModel }
]);
}
}}
>
<Flex alignItems={'center'} h={'38px'}>
<Avatar src={item.avatar} w={['24px', '28px', '32px']}></Avatar>
<Box ml={3} fontWeight={'bold'} fontSize={['md', 'lg', 'xl']}>
{item.name}
</Box>
</Flex>
<Flex justifyContent={'flex-end'} alignItems={'center'} fontSize={'sm'}>
<MyIcon mr={1} name="kbTest" w={'12px'} />
<Box color={'myGray.500'}>{item.vectorModel.name}</Box>
</Flex>
</Card>
);
})()
)}
</ModalBody>
<ModalFooter>
<Button
onClick={() => {
onClose();
onChange(selectedKbList);
}}
>
</Button>
</ModalFooter>
</Flex>
</MyModal>
);
};
export const KbParamsModal = ({
searchEmptyText,
searchLimit,
searchSimilarity,
onClose,
onChange
}: KbParamsType & { onClose: () => void; onChange: (e: KbParamsType) => void }) => {
const [refresh, setRefresh] = useState(false);
const { register, setValue, getValues, handleSubmit } = useForm<KbParamsType>({
defaultValues: {
searchEmptyText,
searchLimit,
searchSimilarity
}
});
return (
<MyModal isOpen={true} onClose={onClose} title={'搜索参数调整'} minW={['90vw', '600px']}>
<Flex flexDirection={'column'}>
<ModalBody>
<Box display={['block', 'flex']} py={5} pt={[0, 5]}>
<Box flex={'0 0 100px'} mb={[8, 0]}>
<MyTooltip
label={'不同索引模型的相似度有区别,请通过搜索测试来选择合适的数值'}
forceShow
>
<QuestionOutlineIcon ml={1} />
</MyTooltip>
</Box>
<MySlider
markList={[
{ label: '0', value: 0 },
{ label: '1', value: 1 }
]}
min={0}
max={1}
step={0.01}
value={getValues('searchSimilarity')}
onChange={(val) => {
setValue('searchSimilarity', val);
setRefresh(!refresh);
}}
/>
</Box>
<Box display={['block', 'flex']} py={8}>
<Box flex={'0 0 100px'} mb={[8, 0]}>
</Box>
<Box flex={1}>
<MySlider
markList={[
{ label: '1', value: 1 },
{ label: '20', value: 20 }
]}
min={1}
max={20}
value={getValues('searchLimit')}
onChange={(val) => {
setValue('searchLimit', val);
setRefresh(!refresh);
}}
/>
</Box>
</Box>
<Box display={['block', 'flex']} pt={3}>
<Box flex={'0 0 100px'} mb={[2, 0]}>
</Box>
<Box flex={1}>
<Textarea
rows={5}
maxLength={500}
placeholder={
'若填写该内容,没有搜索到对应内容时,将直接回复填写的内容。\n为了连贯上下文FastGpt 会取部分上一个聊天的搜索记录作为补充,因此在连续对话时,该功能可能会失效。'
}
{...register('searchEmptyText')}
></Textarea>
</Box>
</Box>
</ModalBody>
<ModalFooter>
<Button variant={'base'} mr={3} onClick={onClose}>
</Button>
<Button
onClick={() => {
onClose();
handleSubmit(onChange)();
}}
>
</Button>
</ModalFooter>
</Flex>
</MyModal>
);
};
export default KBSelectModal;

View File

@@ -1,263 +0,0 @@
import React, { useState } from 'react';
import {
Flex,
Box,
Button,
TableContainer,
Table,
Thead,
Tr,
Th,
Td,
Tbody,
useDisclosure,
ModalFooter,
ModalBody,
FormControl,
Input,
useTheme
} from '@chakra-ui/react';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
import MyIcon from '@/components/Icon';
import { useLoading } from '@/hooks/useLoading';
import { useQuery } from '@tanstack/react-query';
import { getShareChatList, delShareChatById, createShareChat } from '@/api/chat';
import { formatTimeToChatTime, useCopyData } from '@/utils/tools';
import { useForm } from 'react-hook-form';
import { defaultShareChat } from '@/constants/model';
import type { ShareChatEditType } from '@/types/app';
import { useRequest } from '@/hooks/useRequest';
import { formatPrice } from '@/utils/user';
import MyTooltip from '@/components/MyTooltip';
import MyModal from '@/components/MyModal';
import MyRadio from '@/components/Radio';
const Share = ({ appId }: { appId: string }) => {
const { Loading, setIsLoading } = useLoading();
const { copyData } = useCopyData();
const {
isOpen: isOpenCreateShareChat,
onOpen: onOpenCreateShareChat,
onClose: onCloseCreateShareChat
} = useDisclosure();
const {
register: registerShareChat,
getValues: getShareChatValues,
setValue: setShareChatValues,
handleSubmit: submitShareChat,
reset: resetShareChat
} = useForm({
defaultValues: defaultShareChat
});
const {
isFetching,
data: shareChatList = [],
refetch: refetchShareChatList
} = useQuery(['initShareChatList', appId], () => getShareChatList(appId));
const { mutate: onclickCreateShareChat, isLoading: creating } = useRequest({
mutationFn: async (e: ShareChatEditType) =>
createShareChat({
...e,
appId
}),
errorToast: '创建分享链接异常',
onSuccess(id) {
onCloseCreateShareChat();
refetchShareChatList();
const url = `${location.origin}/chat/share?shareId=${id}`;
copyData(url, '创建成功。已复制分享地址,可直接分享使用');
resetShareChat(defaultShareChat);
}
});
return (
<Box position={'relative'} pt={[3, 5, 8]} px={[5, 8]} minH={'50vh'}>
<Flex justifyContent={'space-between'}>
<Box fontWeight={'bold'}>
<MyTooltip
forceShow
label="可以直接分享该模型给其他用户去进行对话对方无需登录即可直接进行对话。注意这个功能会消耗你账号的tokens。请保管好链接和密码。"
>
<QuestionOutlineIcon ml={1} />
</MyTooltip>
</Box>
<Button
variant={'base'}
colorScheme={'myBlue'}
size={['sm', 'md']}
{...(shareChatList.length >= 10
? {
isDisabled: true,
title: '最多创建10组'
}
: {})}
onClick={onOpenCreateShareChat}
>
</Button>
</Flex>
<TableContainer mt={3}>
<Table variant={'simple'} w={'100%'} overflowX={'auto'}>
<Thead>
<Tr>
<Th></Th>
<Th></Th>
<Th>使</Th>
<Th></Th>
</Tr>
</Thead>
<Tbody>
{shareChatList.map((item) => (
<Tr key={item._id}>
<Td>{item.name}</Td>
<Td>{formatPrice(item.total)}</Td>
<Td>{item.lastTime ? formatTimeToChatTime(item.lastTime) : '未使用'}</Td>
<Td display={'flex'} alignItems={'center'}>
<MyTooltip label={'嵌入网页'}>
<MyIcon
mr={4}
name="apiLight"
w={'14px'}
cursor={'pointer'}
_hover={{ color: 'myBlue.600' }}
onClick={() => {
const url = `${location.origin}/chat/share?shareId=${item.shareId}`;
const src = `${location.origin}/js/iframe.js`;
const script = `<script src="${src}" id="fastgpt-iframe" data-src="${url}" data-color="#4e83fd"></script>`;
copyData(script, '已复制嵌入 Script可在应用 HTML 底部嵌入', 3000);
}}
/>
</MyTooltip>
<MyTooltip label={'复制分享链接'}>
<MyIcon
mr={4}
name="copy"
w={'14px'}
cursor={'pointer'}
_hover={{ color: 'myBlue.600' }}
onClick={() => {
const url = `${location.origin}/chat/share?shareId=${item.shareId}`;
copyData(url, '已复制分享链接,可直接分享使用');
}}
/>
</MyTooltip>
<MyTooltip label={'删除链接'}>
<MyIcon
name="delete"
w={'14px'}
cursor={'pointer'}
_hover={{ color: 'red' }}
onClick={async () => {
setIsLoading(true);
try {
await delShareChatById(item._id);
refetchShareChatList();
} catch (error) {
console.log(error);
}
setIsLoading(false);
}}
/>
</MyTooltip>
</Td>
</Tr>
))}
</Tbody>
</Table>
</TableContainer>
{shareChatList.length === 0 && !isFetching && (
<Flex h={'100%'} flexDirection={'column'} alignItems={'center'} pt={'10vh'}>
<MyIcon name="empty" w={'48px'} h={'48px'} color={'transparent'} />
<Box mt={2} color={'myGray.500'}>
</Box>
</Flex>
)}
{/* create shareChat modal */}
<MyModal
isOpen={isOpenCreateShareChat}
onClose={onCloseCreateShareChat}
title={'创建免登录窗口'}
>
<ModalBody>
<FormControl>
<Flex alignItems={'center'}>
<Box flex={'0 0 60px'} w={0}>
:
</Box>
<Input
placeholder="记录名字,仅用于展示"
maxLength={20}
{...registerShareChat('name', {
required: '记录名称不能为空'
})}
/>
</Flex>
</FormControl>
</ModalBody>
<ModalFooter>
<Button variant={'base'} mr={3} onClick={onCloseCreateShareChat}>
</Button>
<Button
isLoading={creating}
onClick={submitShareChat((data) => onclickCreateShareChat(data))}
>
</Button>
</ModalFooter>
</MyModal>
<Loading loading={isFetching} fixed={false} />
</Box>
);
};
enum LinkTypeEnum {
share = 'share',
iframe = 'iframe'
}
const OutLink = ({ appId }: { appId: string }) => {
const theme = useTheme();
const [linkType, setLinkType] = useState<`${LinkTypeEnum}`>(LinkTypeEnum.share);
return (
<Box pt={[1, 5]}>
<Box fontWeight={'bold'} fontSize={['md', 'xl']} mb={2} px={[4, 8]}>
使
</Box>
<Box pb={[5, 7]} px={[4, 8]} borderBottom={theme.borders.base}>
<MyRadio
gridTemplateColumns={['repeat(1,1fr)', 'repeat(auto-fill, minmax(0, 360px))']}
iconSize={'20px'}
list={[
{
icon: 'outlink_share',
title: '免登录窗口',
desc: '分享链接给其他用户,无需登录即可直接进行使用',
value: LinkTypeEnum.share
}
// {
// icon: 'outlink_iframe',
// title: '网页嵌入',
// desc: '嵌入到已有网页中,右下角会生成对话按键',
// value: LinkTypeEnum.iframe
// }
]}
value={linkType}
onChange={(e) => setLinkType(e as `${LinkTypeEnum}`)}
/>
</Box>
{linkType === LinkTypeEnum.share && <Share appId={appId} />}
</Box>
);
};
export default OutLink;

View File

@@ -1,7 +0,0 @@
.intro {
display: -webkit-box;
-webkit-line-clamp: 3;
-webkit-box-orient: vertical;
overflow: hidden;
text-overflow: ellipsis;
}

View File

@@ -1,7 +0,0 @@
.intro {
display: -webkit-box;
-webkit-line-clamp: 3;
-webkit-box-orient: vertical;
overflow: hidden;
text-overflow: ellipsis;
}

View File

@@ -1,40 +0,0 @@
import React, { useEffect } from 'react';
import { Box } from '@chakra-ui/react';
import { serviceSideProps } from '@/utils/i18n';
import Navbar from './components/Navbar';
import Hero from './components/Hero';
import Ability from './components/Ability';
import Choice from './components/Choice';
import Footer from './components/Footer';
const Home = () => {
return (
<Box id="home" bg={'myWhite.600'} h={'100vh'} overflowY={'auto'} overflowX={'hidden'}>
<Box position={'fixed'} zIndex={10} top={0} left={0} right={0}>
<Navbar />
</Box>
<Box maxW={'1200px'} pt={'70px'} m={'auto'}>
<Hero />
<Ability />
<Box my={[4, 6]}>
<Choice />
</Box>
</Box>
<Box bg={'white'}>
<Footer />
</Box>
</Box>
);
};
export async function getServerSideProps(content: any) {
return {
props: {
...(await serviceSideProps(content))
}
};
}
export default Home;

View File

@@ -1,83 +0,0 @@
import React, { useState } from 'react';
import { Box, type BoxProps, Flex, Textarea, useTheme } from '@chakra-ui/react';
import MyRadio from '@/components/Radio/index';
import dynamic from 'next/dynamic';
import ManualImport from './Import/Manual';
const ChunkImport = dynamic(() => import('./Import/Chunk'), {
ssr: true
});
const QAImport = dynamic(() => import('./Import/QA'), {
ssr: true
});
const CsvImport = dynamic(() => import('./Import/Csv'), {
ssr: true
});
enum ImportTypeEnum {
manual = 'manual',
index = 'index',
qa = 'qa',
csv = 'csv'
}
const ImportData = ({ kbId }: { kbId: string }) => {
const theme = useTheme();
const [importType, setImportType] = useState<`${ImportTypeEnum}`>(ImportTypeEnum.manual);
const TitleStyle: BoxProps = {
fontWeight: 'bold',
fontSize: ['md', 'xl'],
mb: [3, 5]
};
return (
<Flex flexDirection={'column'} h={'100%'} pt={[1, 5]}>
<Box {...TitleStyle} px={[4, 8]}>
</Box>
<Box pb={[5, 7]} px={[4, 8]} borderBottom={theme.borders.base}>
<MyRadio
gridTemplateColumns={['repeat(1,1fr)', 'repeat(2, 350px)']}
list={[
{
icon: 'manualImport',
title: '手动输入',
desc: '手动输入问答对,是最精准的数据',
value: ImportTypeEnum.manual
},
{
icon: 'indexImport',
title: '直接分段',
desc: '选择文本文件,直接将其按分段进行处理',
value: ImportTypeEnum.index
},
{
icon: 'qaImport',
title: 'QA拆分',
desc: '选择文本文件,让大模型自动生成问答对',
value: ImportTypeEnum.qa
},
{
icon: 'csvImport',
title: 'CSV 导入',
desc: '批量导入问答对,是最精准的数据',
value: ImportTypeEnum.csv
}
]}
value={importType}
onChange={(e) => setImportType(e as `${ImportTypeEnum}`)}
/>
</Box>
<Box flex={'1 0 0'} h={0}>
{importType === ImportTypeEnum.manual && <ManualImport kbId={kbId} />}
{importType === ImportTypeEnum.index && <ChunkImport kbId={kbId} />}
{importType === ImportTypeEnum.qa && <QAImport kbId={kbId} />}
{importType === ImportTypeEnum.csv && <CsvImport kbId={kbId} />}
</Box>
</Flex>
);
};
export default ImportData;

View File

@@ -1,431 +0,0 @@
import React, { useState, useCallback, useMemo } from 'react';
import {
Box,
Flex,
Button,
useTheme,
NumberInput,
NumberInputField,
NumberInputStepper,
NumberIncrementStepper,
NumberDecrementStepper,
Image
} from '@chakra-ui/react';
import { useToast } from '@/hooks/useToast';
import { useConfirm } from '@/hooks/useConfirm';
import { useRouter } from 'next/router';
import { useMutation } from '@tanstack/react-query';
import { postKbDataFromList } from '@/api/plugins/kb';
import { splitText2Chunks } from '@/utils/file';
import { getErrText } from '@/utils/tools';
import { formatPrice } from '@/utils/user';
import MyIcon from '@/components/Icon';
import CloseIcon from '@/components/Icon/close';
import DeleteIcon, { hoverDeleteStyles } from '@/components/Icon/delete';
import MyTooltip from '@/components/MyTooltip';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
import { TrainingModeEnum } from '@/constants/plugin';
import FileSelect, { type FileItemType } from './FileSelect';
import { useUserStore } from '@/store/user';
const fileExtension = '.txt, .doc, .docx, .pdf, .md';
const ChunkImport = ({ kbId }: { kbId: string }) => {
const { kbDetail } = useUserStore();
const vectorModel = kbDetail.vectorModel;
const unitPrice = vectorModel?.price || 0.2;
const theme = useTheme();
const router = useRouter();
const { toast } = useToast();
const [chunkLen, setChunkLen] = useState(vectorModel?.defaultToken || 300);
const [showRePreview, setShowRePreview] = useState(false);
const [files, setFiles] = useState<FileItemType[]>([]);
const [previewFile, setPreviewFile] = useState<FileItemType>();
const [successChunks, setSuccessChunks] = useState(0);
const totalChunk = useMemo(
() => files.reduce((sum, file) => sum + file.chunks.length, 0),
[files]
);
const emptyFiles = useMemo(() => files.length === 0, [files]);
// price count
const price = useMemo(() => {
return formatPrice(files.reduce((sum, file) => sum + file.tokens, 0) * unitPrice);
}, [files, unitPrice]);
const { openConfirm, ConfirmModal } = useConfirm({
content: `该任务无法终止,需要一定时间生成索引,请确认导入。如果余额不足,未完成的任务会被暂停,充值后可继续进行。`
});
const { mutate: onclickUpload, isLoading: uploading } = useMutation({
mutationFn: async () => {
const chunks = files.map((file) => file.chunks).flat();
// subsection import
let success = 0;
const step = 500;
for (let i = 0; i < chunks.length; i += step) {
const { insertLen } = await postKbDataFromList({
kbId,
data: chunks.slice(i, i + step),
mode: TrainingModeEnum.index
});
success += insertLen;
setSuccessChunks(success);
}
toast({
title: `去重后共导入 ${success} 条数据,请耐心等待训练.`,
status: 'success'
});
router.replace({
query: {
kbId,
currentTab: 'data'
}
});
},
onError(err) {
toast({
title: getErrText(err, '导入文件失败'),
status: 'error'
});
}
});
const onRePreview = useCallback(async () => {
try {
setFiles((state) =>
state.map((file) => {
const splitRes = splitText2Chunks({
text: file.text,
maxLen: chunkLen
});
return {
...file,
tokens: splitRes.tokens,
chunks: splitRes.chunks.map((chunk) => ({
q: chunk,
a: '',
source: file.filename
}))
};
})
);
setPreviewFile(undefined);
setShowRePreview(false);
} catch (error) {
toast({
status: 'warning',
title: getErrText(error, '文本分段异常')
});
}
}, [chunkLen, toast]);
const filenameStyles = {
className: 'textEllipsis',
maxW: '400px'
};
return (
<Box display={['block', 'flex']} h={['auto', '100%']} overflow={'overlay'}>
<Flex
flexDirection={'column'}
flex={'1 0 0'}
h={'100%'}
minW={['auto', '400px']}
w={['100%', 0]}
p={[4, 8]}
>
<FileSelect
fileExtension={fileExtension}
onPushFiles={(files) => {
setFiles((state) => files.concat(state));
}}
chunkLen={chunkLen}
py={emptyFiles ? '100px' : 5}
/>
{!emptyFiles && (
<>
<Box py={4} px={2} minH={['auto', '100px']} maxH={'400px'} overflow={'auto'}>
{files.map((item) => (
<Flex
key={item.id}
w={'100%'}
_notLast={{ mb: 5 }}
px={5}
py={2}
boxShadow={'1px 1px 5px rgba(0,0,0,0.15)'}
borderRadius={'md'}
cursor={'pointer'}
position={'relative'}
alignItems={'center'}
_hover={{
bg: 'myBlue.100',
'& .delete': {
display: 'block'
}
}}
onClick={() => setPreviewFile(item)}
>
<Image src={item.icon} w={'16px'} alt={''} />
<Box ml={2} flex={'1 0 0'} pr={3} {...filenameStyles}>
{item.filename}
</Box>
<MyIcon
position={'absolute'}
right={3}
className="delete"
name={'delete'}
w={'16px'}
_hover={{ color: 'red.600' }}
display={['block', 'none']}
onClick={(e) => {
e.stopPropagation();
setFiles((state) => state.filter((file) => file.id !== item.id));
}}
/>
</Flex>
))}
</Box>
{/* chunk size */}
<Flex py={5} alignItems={'center'}>
<Box>
<MyTooltip
label={
'按结束标点符号进行分段。前后段落会有 30% 的内容重叠。\n中文文档建议不要超过800英文不要超过1500'
}
forceShow
>
<QuestionOutlineIcon ml={1} />
</MyTooltip>
</Box>
<Box
flex={1}
css={{
'& > span': {
display: 'block'
}
}}
>
<MyTooltip label={`范围: 100~${kbDetail.vectorModel.maxToken}`}>
<NumberInput
ml={4}
defaultValue={chunkLen}
min={100}
max={kbDetail.vectorModel.maxToken}
step={10}
onChange={(e) => {
setChunkLen(+e);
setShowRePreview(true);
}}
>
<NumberInputField />
<NumberInputStepper>
<NumberIncrementStepper />
<NumberDecrementStepper />
</NumberInputStepper>
</NumberInput>
</MyTooltip>
</Box>
</Flex>
{/* price */}
<Flex py={5} alignItems={'center'}>
<Box>
<MyTooltip
label={`索引生成计费为: ${formatPrice(unitPrice, 1000)}/1k tokens`}
forceShow
>
<QuestionOutlineIcon ml={1} />
</MyTooltip>
</Box>
<Box ml={4}>{price}</Box>
</Flex>
<Flex mt={3}>
{showRePreview && (
<Button variant={'base'} mr={4} onClick={onRePreview}>
</Button>
)}
<Button isDisabled={uploading} onClick={openConfirm(onclickUpload)}>
{uploading ? (
<Box>{Math.round((successChunks / totalChunk) * 100)}%</Box>
) : (
'确认导入'
)}
</Button>
</Flex>
</>
)}
</Flex>
{!emptyFiles && (
<Box flex={'2 0 0'} w={['100%', 0]} h={'100%'}>
{previewFile ? (
<Box
position={'relative'}
display={['block', 'flex']}
h={'100%'}
flexDirection={'column'}
pt={[4, 8]}
bg={'myWhite.400'}
>
<Box px={[4, 8]} fontSize={['lg', 'xl']} fontWeight={'bold'} {...filenameStyles}>
{previewFile.filename}
</Box>
<CloseIcon
position={'absolute'}
right={[4, 8]}
top={4}
onClick={() => setPreviewFile(undefined)}
/>
<Box
flex={'1 0 0'}
h={['auto', 0]}
overflow={'overlay'}
px={[4, 8]}
my={4}
contentEditable
dangerouslySetInnerHTML={{ __html: previewFile.text }}
fontSize={'sm'}
whiteSpace={'pre-wrap'}
wordBreak={'break-all'}
onBlur={(e) => {
// @ts-ignore
const val = e.target.innerText;
setShowRePreview(true);
setFiles((state) =>
state.map((file) =>
file.id === previewFile.id
? {
...file,
text: val
}
: file
)
);
}}
/>
</Box>
) : (
<Box h={'100%'} pt={[4, 8]} overflow={'overlay'}>
<Flex px={[4, 8]} alignItems={'center'}>
<Box fontSize={['lg', 'xl']} fontWeight={'bold'}>
({totalChunk})
</Box>
{totalChunk > 100 && (
<Box ml={2} fontSize={'sm'} color={'myhGray.500'}>
</Box>
)}
</Flex>
<Box px={[4, 8]} overflow={'overlay'}>
{files.map((file) =>
file.chunks.slice(0, 50).map((chunk, i) => (
<Box
key={i}
py={4}
bg={'myWhite.500'}
my={2}
borderRadius={'md'}
fontSize={'sm'}
_hover={{ ...hoverDeleteStyles }}
>
<Flex mb={1} px={4} userSelect={'none'}>
<Box
flexShrink={0}
px={3}
py={'1px'}
border={theme.borders.base}
borderRadius={'md'}
>
# {i + 1}
</Box>
<Box ml={2} fontSize={'sm'} color={'myhGray.500'} {...filenameStyles}>
{file.filename}
</Box>
<Box flex={1} />
<DeleteIcon
onClick={() => {
setFiles((state) =>
state.map((stateFile) =>
stateFile.id === file.id
? {
...file,
chunks: [
...file.chunks.slice(0, i),
...file.chunks.slice(i + 1)
]
}
: stateFile
)
);
}}
/>
</Flex>
<Box
px={4}
fontSize={'sm'}
whiteSpace={'pre-wrap'}
wordBreak={'break-all'}
contentEditable
dangerouslySetInnerHTML={{ __html: chunk.q }}
onBlur={(e) => {
// @ts-ignore
const val = e.target.innerText;
/* delete file */
if (val === '') {
setFiles((state) =>
state.map((stateFile) =>
stateFile.id === file.id
? {
...file,
chunks: [
...file.chunks.slice(0, i),
...file.chunks.slice(i + 1)
]
}
: stateFile
)
);
} else {
// update file
setFiles((stateFiles) =>
stateFiles.map((stateFile) =>
file.id === stateFile.id
? {
...stateFile,
chunks: stateFile.chunks.map((chunk, index) => ({
...chunk,
q: i === index ? val : chunk.q
}))
}
: stateFile
)
);
}
}}
/>
</Box>
))
)}
</Box>
</Box>
)}
</Box>
)}
<ConfirmModal />
</Box>
);
};
export default ChunkImport;

View File

@@ -1,223 +0,0 @@
import React, { useState, useCallback, 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';
import FileSelect, { type FileItemType } from './FileSelect';
import { useRouter } from 'next/router';
import { useUserStore } from '@/store/user';
const fileExtension = '.csv';
const CsvImport = ({ kbId }: { kbId: string }) => {
const { kbDetail } = useUserStore();
const maxToken = kbDetail.vectorModel?.maxToken || 2000;
const theme = useTheme();
const router = useRouter();
const { toast } = useToast();
const [files, setFiles] = useState<FileItemType[]>([]);
const [successChunks, setSuccessChunks] = useState(0);
const totalChunk = useMemo(
() => files.reduce((sum, file) => sum + file.chunks.length, 0),
[files]
);
const emptyFiles = useMemo(() => files.length === 0, [files]);
const { openConfirm, ConfirmModal } = useConfirm({
content: `该任务无法终止,需要一定时间生成索引,请确认导入。如果余额不足,未完成的任务会被暂停,充值后可继续进行。`
});
const { mutate: onclickUpload, isLoading: uploading } = useMutation({
mutationFn: async () => {
const chunks = files
.map((file) => file.chunks)
.flat()
.filter((item) => item?.q);
const filterChunks = chunks.filter((item) => item.q.length < maxToken);
if (filterChunks.length !== chunks.length) {
toast({
title: `${chunks.length - filterChunks.length}条数据超出长度,已被过滤`,
status: 'info'
});
}
// subsection import
let success = 0;
const step = 500;
for (let i = 0; i < filterChunks.length; i += step) {
const { insertLen } = await postKbDataFromList({
kbId,
data: filterChunks.slice(i, i + step),
mode: TrainingModeEnum.index
});
success += insertLen;
setSuccessChunks(success);
}
toast({
title: `去重后共导入 ${success} 条数据,请耐心等待训练.`,
status: 'success'
});
router.replace({
query: {
kbId,
currentTab: 'data'
}
});
},
onError(err) {
toast({
title: getErrText(err, '导入文件失败'),
status: 'error'
});
}
});
const filenameStyles = {
className: 'textEllipsis',
maxW: '400px'
};
return (
<Box display={['block', 'flex']} h={['auto', '100%']} overflow={'overlay'}>
<Flex
flexDirection={'column'}
flex={'1 0 0'}
h={'100%'}
minW={['auto', '400px']}
w={['100%', 0]}
p={[4, 8]}
>
<FileSelect
fileExtension={fileExtension}
tipText={
'file.If the imported file is garbled, please convert CSV to UTF-8 encoding format'
}
onPushFiles={(files) => setFiles((state) => files.concat(state))}
showUrlFetch={false}
showCreateFile={false}
py={emptyFiles ? '100px' : 5}
isCsv
/>
{!emptyFiles && (
<>
<Box py={4} minH={['auto', '100px']} px={2} maxH={'400px'} overflow={'auto'}>
{files.map((item) => (
<Flex
key={item.id}
w={'100%'}
_notLast={{ mb: 5 }}
px={5}
py={2}
boxShadow={'1px 1px 5px rgba(0,0,0,0.15)'}
borderRadius={'md'}
position={'relative'}
alignItems={'center'}
_hover={{ ...hoverDeleteStyles }}
>
<Image src={'/imgs/files/csv.svg'} w={'16px'} alt={''} />
<Box ml={2} flex={'1 0 0'} pr={3} {...filenameStyles}>
{item.filename}
</Box>
<MyIcon
position={'absolute'}
right={3}
className="delete"
name={'delete'}
w={'16px'}
_hover={{ color: 'red.600' }}
display={['block', 'none']}
onClick={(e) => {
e.stopPropagation();
setFiles((state) => state.filter((file) => file.id !== item.id));
}}
/>
</Flex>
))}
</Box>
<Flex mt={3}>
<Button isDisabled={uploading} onClick={openConfirm(onclickUpload)}>
{uploading ? (
<Box>{Math.round((successChunks / totalChunk) * 100)}%</Box>
) : (
'确认导入'
)}
</Button>
</Flex>
</>
)}
</Flex>
{!emptyFiles && (
<Box flex={'2 0 0'} w={['100%', 0]} h={'100%'} pt={[4, 8]} overflow={'overlay'}>
<Flex px={[4, 8]} alignItems={'center'}>
<Box fontSize={['lg', 'xl']} fontWeight={'bold'}>
({totalChunk})
</Box>
{totalChunk > 100 && (
<Box ml={2} fontSize={'sm'} color={'myhGray.500'}>
</Box>
)}
</Flex>
<Box px={[4, 8]} overflow={'overlay'}>
{files.map((file) =>
file.chunks.slice(0, 100).map((item, i) => (
<Box
key={i}
py={4}
bg={'myWhite.500'}
my={2}
borderRadius={'md'}
fontSize={'sm'}
_hover={{ ...hoverDeleteStyles }}
>
<Flex mb={1} px={4} userSelect={'none'}>
<Box px={3} py={'1px'} border={theme.borders.base} borderRadius={'md'}>
# {i + 1}
</Box>
{item.source && <Box ml={1}>({item.source})</Box>}
<Box flex={1} />
<DeleteIcon
onClick={() => {
setFiles((state) =>
state.map((stateFile) =>
stateFile.id === file.id
? {
...file,
chunks: [...file.chunks.slice(0, i), ...file.chunks.slice(i + 1)]
}
: stateFile
)
);
}}
/>
</Flex>
<Box px={4} fontSize={'sm'} whiteSpace={'pre-wrap'} wordBreak={'break-all'}>
{`q: ${item.q}\na: ${item.a}`}
</Box>
</Box>
))
)}
</Box>
</Box>
)}
<ConfirmModal />
</Box>
);
};
export default CsvImport;

View File

@@ -1,122 +0,0 @@
import React, { useState } from 'react';
import { Box, Textarea, Button, Flex } from '@chakra-ui/react';
import { useForm } from 'react-hook-form';
import { useToast } from '@/hooks/useToast';
import { useRequest } from '@/hooks/useRequest';
import { getErrText } from '@/utils/tools';
import { postKbDataFromList } from '@/api/plugins/kb';
import { TrainingModeEnum } from '@/constants/plugin';
import { useUserStore } from '@/store/user';
import MyTooltip from '@/components/MyTooltip';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
type ManualFormType = { q: string; a: string };
const ManualImport = ({ kbId }: { kbId: string }) => {
const { kbDetail } = useUserStore();
const maxToken = kbDetail.vectorModel?.maxToken || 2000;
const { register, handleSubmit, reset } = useForm({
defaultValues: { q: '', a: '' }
});
const { toast } = useToast();
const [qLen, setQLen] = useState(0);
const { mutate: onImportData, isLoading } = useRequest({
mutationFn: async (e: ManualFormType) => {
if (e.a.length + e.q.length >= 3000) {
toast({
title: '总长度超长了',
status: 'warning'
});
return;
}
try {
const data = {
a: e.a,
q: e.q,
source: '手动录入'
};
const { insertLen } = await postKbDataFromList({
kbId,
mode: TrainingModeEnum.index,
data: [data]
});
if (insertLen === 0) {
toast({
title: '已存在完全一致的数据',
status: 'warning'
});
} else {
toast({
title: '导入数据成功,需要一段时间训练',
status: 'success'
});
reset({
a: '',
q: ''
});
}
} catch (err: any) {
toast({
title: getErrText(err, '出现了点意外~'),
status: 'error'
});
}
}
});
return (
<Box p={[4, 8]} h={'100%'} overflow={'overlay'}>
<Box display={'flex'} flexDirection={['column', 'row']}>
<Box flex={1} mr={[0, 4]} mb={[4, 0]} h={['50%', '100%']} position={'relative'}>
<Flex>
<Box h={'30px'}>{'匹配的知识点'}</Box>
<MyTooltip label={'被向量化的部分,通常是问题,也可以是一段陈述描述'}>
<QuestionOutlineIcon ml={1} />
</MyTooltip>
</Flex>
<Textarea
placeholder={`匹配的知识点。这部分内容会被搜索,请把控内容的质量。最多 ${maxToken} 字。`}
maxLength={maxToken}
h={['250px', '500px']}
{...register(`q`, {
required: true,
onChange(e) {
setQLen(e.target.value.length);
}
})}
/>
<Box position={'absolute'} color={'myGray.500'} right={5} bottom={3} zIndex={99}>
{qLen}
</Box>
</Box>
<Box flex={1} h={['50%', '100%']}>
<Flex>
<Box h={'30px'}>{'补充知识'}</Box>
<MyTooltip
label={'匹配的知识点被命中后,这部分内容会随匹配知识点一起注入模型,引导模型回答'}
>
<QuestionOutlineIcon ml={1} />
</MyTooltip>
</Flex>
<Textarea
placeholder={
'补充知识。这部分内容不会被搜索,但会作为"匹配的知识点"的内容补充,你可以讲一些细节的内容填写在这里。总和最多 3000 字。'
}
h={['250px', '500px']}
maxLength={3000}
{...register('a')}
/>
</Box>
</Box>
<Button mt={5} isLoading={isLoading} onClick={handleSubmit((data) => onImportData(data))}>
</Button>
</Box>
);
};
export default React.memo(ManualImport);

View File

@@ -1,392 +0,0 @@
import React, { useState, useCallback, useMemo } from 'react';
import { Box, Flex, Button, useTheme, Image, Input } 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 { splitText2Chunks } from '@/utils/file';
import { getErrText } from '@/utils/tools';
import { formatPrice } from '@/utils/user';
import { qaModel } from '@/store/static';
import MyIcon from '@/components/Icon';
import CloseIcon from '@/components/Icon/close';
import DeleteIcon, { hoverDeleteStyles } from '@/components/Icon/delete';
import MyTooltip from '@/components/MyTooltip';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
import { TrainingModeEnum } from '@/constants/plugin';
import FileSelect, { type FileItemType } from './FileSelect';
import { useRouter } from 'next/router';
const fileExtension = '.txt, .doc, .docx, .pdf, .md';
const QAImport = ({ kbId }: { kbId: string }) => {
const unitPrice = qaModel.price || 3;
const chunkLen = qaModel.maxToken * 0.45;
const theme = useTheme();
const router = useRouter();
const { toast } = useToast();
const [files, setFiles] = useState<FileItemType[]>([]);
const [showRePreview, setShowRePreview] = useState(false);
const [previewFile, setPreviewFile] = useState<FileItemType>();
const [successChunks, setSuccessChunks] = useState(0);
const [prompt, setPrompt] = useState('');
const totalChunk = useMemo(
() => files.reduce((sum, file) => sum + file.chunks.length, 0),
[files]
);
const emptyFiles = useMemo(() => files.length === 0, [files]);
// price count
const price = useMemo(() => {
return formatPrice(files.reduce((sum, file) => sum + file.tokens, 0) * unitPrice * 1.3);
}, [files, unitPrice]);
const { openConfirm, ConfirmModal } = useConfirm({
content: `该任务无法终止!导入后会自动调用大模型生成问答对,会有一些细节丢失,请确认!如果余额不足,未完成的任务会被暂停。`
});
const { mutate: onclickUpload, isLoading: uploading } = useMutation({
mutationFn: async () => {
const chunks = files.map((file) => file.chunks).flat();
// subsection import
let success = 0;
const step = 300;
for (let i = 0; i < chunks.length; i += step) {
const { insertLen } = await postKbDataFromList({
kbId,
data: chunks.slice(i, i + step),
mode: TrainingModeEnum.qa,
prompt: prompt || '下面是一段长文本'
});
success += insertLen;
setSuccessChunks(success);
}
toast({
title: `共导入 ${success} 条数据,请耐心等待训练.`,
status: 'success'
});
router.replace({
query: {
kbId,
currentTab: 'data'
}
});
},
onError(err) {
toast({
title: getErrText(err, '导入文件失败'),
status: 'error'
});
}
});
const onRePreview = useCallback(async () => {
try {
setFiles((state) =>
state.map((file) => {
const splitRes = splitText2Chunks({
text: file.text,
maxLen: chunkLen
});
return {
...file,
tokens: splitRes.tokens,
chunks: splitRes.chunks.map((chunk) => ({
q: chunk,
a: '',
source: file.filename
}))
};
})
);
setPreviewFile(undefined);
setShowRePreview(false);
} catch (error) {
toast({
status: 'warning',
title: getErrText(error, '文本分段异常')
});
}
}, [chunkLen, toast]);
const filenameStyles = {
className: 'textEllipsis',
maxW: '400px'
};
return (
<Box display={['block', 'flex']} h={['auto', '100%']} overflow={'overlay'}>
<Flex
flexDirection={'column'}
flex={'1 0 0'}
h={'100%'}
minW={['auto', '400px']}
w={['100%', 0]}
p={[4, 8]}
>
<FileSelect
fileExtension={fileExtension}
onPushFiles={(files) => {
setFiles((state) => files.concat(state));
}}
chunkLen={chunkLen}
py={emptyFiles ? '100px' : 5}
/>
{!emptyFiles && (
<>
<Box py={4} px={2} minH={['auto', '100px']} maxH={'400px'} overflow={'auto'}>
{files.map((item) => (
<Flex
key={item.id}
w={'100%'}
_notLast={{ mb: 5 }}
px={5}
py={2}
boxShadow={'1px 1px 5px rgba(0,0,0,0.15)'}
borderRadius={'md'}
cursor={'pointer'}
position={'relative'}
alignItems={'center'}
_hover={{
bg: 'myBlue.100',
'& .delete': {
display: 'block'
}
}}
onClick={() => setPreviewFile(item)}
>
<Image src={item.icon} w={'16px'} alt={''} />
<Box ml={2} flex={'1 0 0'} pr={3} {...filenameStyles}>
{item.filename}
</Box>
<MyIcon
position={'absolute'}
right={3}
className="delete"
name={'delete'}
w={'16px'}
_hover={{ color: 'red.600' }}
display={['block', 'none']}
onClick={(e) => {
e.stopPropagation();
setFiles((state) => state.filter((file) => file.id !== item.id));
}}
/>
</Flex>
))}
</Box>
{/* prompt */}
<Box py={5}>
<Box mb={2}>
QA {' '}
<MyTooltip
label={`可输入关于文件内容的范围介绍,例如:\n1. Laf 的介绍\n2. xxx的简历\n最终会补全为: 关于{输入的内容}`}
forceShow
>
<QuestionOutlineIcon ml={1} />
</MyTooltip>
</Box>
<Flex alignItems={'center'} fontSize={'sm'}>
<Box mr={2}></Box>
<Input
flex={1}
placeholder={'Laf 云函数的介绍'}
bg={'myWhite.500'}
defaultValue={prompt}
onBlur={(e) => (e.target.value ? setPrompt(`关于"${e.target.value}"`) : '')}
/>
</Flex>
</Box>
{/* price */}
<Flex py={5} alignItems={'center'}>
<Box>
<MyTooltip
label={`索引生成计费为: ${formatPrice(unitPrice, 1000)}/1k tokens`}
forceShow
>
<QuestionOutlineIcon ml={1} />
</MyTooltip>
</Box>
<Box ml={4}>{price}</Box>
</Flex>
<Flex mt={3}>
{showRePreview && (
<Button variant={'base'} mr={4} onClick={onRePreview}>
</Button>
)}
<Button isDisabled={uploading} onClick={openConfirm(onclickUpload)}>
{uploading ? (
<Box>{Math.round((successChunks / totalChunk) * 100)}%</Box>
) : (
'确认导入'
)}
</Button>
</Flex>
</>
)}
</Flex>
{!emptyFiles && (
<Box flex={'2 0 0'} w={['100%', 0]} h={'100%'}>
{previewFile ? (
<Box
position={'relative'}
display={['block', 'flex']}
h={'100%'}
flexDirection={'column'}
pt={[4, 8]}
bg={'myWhite.400'}
>
<Box px={[4, 8]} fontSize={['lg', 'xl']} fontWeight={'bold'} {...filenameStyles}>
{previewFile.filename}
</Box>
<CloseIcon
position={'absolute'}
right={[4, 8]}
top={4}
onClick={() => setPreviewFile(undefined)}
/>
<Box
flex={'1 0 0'}
h={['auto', 0]}
overflow={'overlay'}
px={[4, 8]}
my={4}
contentEditable
dangerouslySetInnerHTML={{ __html: previewFile.text }}
fontSize={'sm'}
whiteSpace={'pre-wrap'}
wordBreak={'break-all'}
onBlur={(e) => {
// @ts-ignore
const val = e.target.innerText;
setShowRePreview(true);
setFiles((state) =>
state.map((file) =>
file.id === previewFile.id
? {
...file,
text: val
}
: file
)
);
}}
/>
</Box>
) : (
<Box h={'100%'} pt={[4, 8]} overflow={'overlay'}>
<Flex px={[4, 8]} alignItems={'center'}>
<Box fontSize={['lg', 'xl']} fontWeight={'bold'}>
({totalChunk})
</Box>
{totalChunk > 100 && (
<Box ml={2} fontSize={'sm'} color={'myhGray.500'}>
</Box>
)}
</Flex>
<Box px={[4, 8]} overflow={'overlay'}>
{files.map((file) =>
file.chunks.slice(0, 30).map((chunk, i) => (
<Box
key={i}
py={4}
bg={'myWhite.500'}
my={2}
borderRadius={'md'}
fontSize={'sm'}
_hover={{ ...hoverDeleteStyles }}
>
<Flex mb={1} px={4} userSelect={'none'}>
<Box px={3} py={'1px'} border={theme.borders.base} borderRadius={'md'}>
# {i + 1}
</Box>
<Box ml={2} fontSize={'sm'} color={'myhGray.500'} {...filenameStyles}>
{file.filename}
</Box>
<Box flex={1} />
<DeleteIcon
onClick={() => {
setFiles((state) =>
state.map((stateFile) =>
stateFile.id === file.id
? {
...file,
chunks: [
...file.chunks.slice(0, i),
...file.chunks.slice(i + 1)
]
}
: stateFile
)
);
}}
/>
</Flex>
<Box
px={4}
fontSize={'sm'}
whiteSpace={'pre-wrap'}
wordBreak={'break-all'}
contentEditable
dangerouslySetInnerHTML={{ __html: chunk.q }}
onBlur={(e) => {
// @ts-ignore
const val = e.target.innerText;
/* delete file */
if (val === '') {
setFiles((state) =>
state.map((stateFile) =>
stateFile.id === file.id
? {
...file,
chunks: [
...file.chunks.slice(0, i),
...file.chunks.slice(i + 1)
]
}
: stateFile
)
);
} else {
// update file
setFiles((stateFiles) =>
stateFiles.map((stateFile) =>
file.id === stateFile.id
? {
...stateFile,
chunks: stateFile.chunks.map((chunk, index) => ({
...chunk,
q: i === index ? val : chunk.q
}))
}
: stateFile
)
);
}
}}
/>
</Box>
))
)}
</Box>
</Box>
)}
</Box>
)}
<ConfirmModal />
</Box>
);
};
export default QAImport;

View File

@@ -1,235 +0,0 @@
import React, { useState, useCallback } from 'react';
import { Box, Flex, Button, Textarea, IconButton } from '@chakra-ui/react';
import { useForm } from 'react-hook-form';
import { insertData2Kb, putKbDataById, delOneKbDataByDataId } from '@/api/plugins/kb';
import { useToast } from '@/hooks/useToast';
import { getErrText } from '@/utils/tools';
import MyIcon from '@/components/Icon';
import MyModal from '@/components/MyModal';
import MyTooltip from '@/components/MyTooltip';
import { QuestionOutlineIcon } from '@chakra-ui/icons';
import { useUserStore } from '@/store/user';
import { useQuery } from '@tanstack/react-query';
export type FormData = { dataId?: string; a: string; q: string; source?: string };
const InputDataModal = ({
onClose,
onSuccess,
onDelete,
kbId,
defaultValues = {
a: '',
q: ''
}
}: {
onClose: () => void;
onSuccess: (data: FormData) => void;
onDelete?: () => void;
kbId: string;
defaultValues?: FormData;
}) => {
const [loading, setLoading] = useState(false);
const { toast } = useToast();
const { kbDetail, getKbDetail } = useUserStore();
const { register, handleSubmit, reset } = useForm<FormData>({
defaultValues
});
const maxToken = kbDetail.vectorModel?.maxToken || 2000;
/**
* 确认导入新数据
*/
const sureImportData = useCallback(
async (e: FormData) => {
if (e.q.length >= maxToken) {
toast({
title: '总长度超长了',
status: 'warning'
});
return;
}
setLoading(true);
try {
const data = {
dataId: '',
a: e.a,
q: e.q,
source: '手动录入'
};
data.dataId = await insertData2Kb({
kbId,
data
});
toast({
title: '导入数据成功,需要一段时间训练',
status: 'success'
});
reset({
a: '',
q: ''
});
onSuccess(data);
} catch (err: any) {
toast({
title: getErrText(err, '出现了点意外~'),
status: 'error'
});
}
setLoading(false);
},
[kbId, maxToken, onSuccess, reset, toast]
);
const updateData = useCallback(
async (e: FormData) => {
if (!e.dataId) return;
if (e.a !== defaultValues.a || e.q !== defaultValues.q) {
setLoading(true);
try {
const data = {
dataId: e.dataId,
kbId,
a: e.a,
q: e.q === defaultValues.q ? '' : e.q
};
await putKbDataById(data);
onSuccess(data);
} catch (err) {
toast({
status: 'error',
title: getErrText(err, '更新数据失败')
});
}
setLoading(false);
}
toast({
title: '修改数据成功',
status: 'success'
});
onClose();
},
[defaultValues.a, defaultValues.q, kbId, onClose, onSuccess, toast]
);
useQuery(['getKbDetail'], () => {
if (kbDetail._id === kbId) return null;
return getKbDetail(kbId);
});
return (
<MyModal
isOpen={true}
onClose={onClose}
isCentered
title={defaultValues.dataId ? '变更数据' : '手动导入数据'}
w={'90vw'}
maxW={'90vw'}
h={'90vh'}
>
<Flex flexDirection={'column'} h={'100%'}>
<Box
display={'flex'}
flexDirection={['column', 'row']}
flex={'1 0 0'}
h={['100%', 0]}
overflow={'overlay'}
px={6}
pb={2}
>
<Box flex={1} mr={[0, 4]} mb={[4, 0]} h={['50%', '100%']}>
<Flex>
<Box h={'30px'}>{'匹配的知识点'}</Box>
<MyTooltip label={'被向量化的部分,通常是问题,也可以是一段陈述描述'}>
<QuestionOutlineIcon ml={1} />
</MyTooltip>
</Flex>
<Textarea
placeholder={`匹配的知识点。这部分内容会被搜索,请把控内容的质量,最多 ${maxToken} 字。`}
maxLength={maxToken}
resize={'none'}
h={'calc(100% - 30px)'}
{...register(`q`, {
required: true
})}
/>
</Box>
<Box flex={1} h={['50%', '100%']}>
<Flex>
<Box h={'30px'}>{'补充知识'}</Box>
<MyTooltip
label={'匹配的知识点被命中后,这部分内容会随匹配知识点一起注入模型,引导模型回答'}
>
<QuestionOutlineIcon ml={1} />
</MyTooltip>
</Flex>
<Textarea
placeholder={
'补充知识。这部分内容不会被搜索,但会作为"匹配的知识点"的内容补充,你可以讲一些细节的内容填写在这里。总和最多 3000 字。'
}
maxLength={3000}
resize={'none'}
h={'calc(100% - 30px)'}
{...register('a')}
/>
</Box>
</Box>
<Flex px={6} pt={2} pb={4} alignItems={'center'}>
<Box flex={1}>
{defaultValues.dataId && onDelete && (
<IconButton
variant={'outline'}
icon={<MyIcon name={'delete'} w={'16px'} h={'16px'} />}
aria-label={''}
isLoading={loading}
size={'sm'}
_hover={{
color: 'red.600',
borderColor: 'red.600'
}}
onClick={async () => {
if (!onDelete || !defaultValues.dataId) return;
try {
await delOneKbDataByDataId(defaultValues.dataId);
onDelete();
onClose();
toast({
status: 'success',
title: '记录已删除'
});
} catch (error) {
toast({
status: 'warning',
title: getErrText(error)
});
console.log(error);
}
}}
/>
)}
</Box>
<Button variant={'base'} mr={3} isLoading={loading} onClick={onClose}>
</Button>
<Button
isLoading={loading}
onClick={handleSubmit(defaultValues.dataId ? updateData : sureImportData)}
>
{defaultValues.dataId ? '确认变更' : '确认导入'}
</Button>
</Flex>
</Flex>
</MyModal>
);
};
export default InputDataModal;

Some files were not shown because too many files have changed in this diff Show More