feat: bump version to 0.3.4 (#406)

Co-authored-by: StyleZhang <jasonapring2015@outlook.com>
Co-authored-by: JzoNg <jzongcode@gmail.com>

.github/ISSUE_TEMPLATE/🐛-bug-report.md

@@ -0,0 +1,32 @@
+---
+name: "\U0001F41B Bug report"
+about: Create a report to help us improve
+title: ''
+labels: bug
+assignees: ''
+
+---
+
+<!--
+  Please provide a clear and concise description of what the bug is. Include
+  screenshots if needed. Please test using the latest version of the relevant
+  Dify packages to make sure your issue has not already been fixed.
+-->
+
+Dify version: Cloud | Self Host
+
+## Steps To Reproduce
+<!--
+  Your bug will get fixed much faster if we can run your code and it doesn't
+  have dependencies other than Dify. Issues without reproduction steps or
+  code examples may be immediately closed as not actionable.
+-->
+
+1.
+2.
+
+
+## The current behavior
+
+
+## The expected behavior

.github/ISSUE_TEMPLATE/🚀-feature-request.md

@@ -0,0 +1,20 @@
+---
+name: "\U0001F680 Feature request"
+about: Suggest an idea for this project
+title: ''
+labels: enhancement
+assignees: ''
+
+---
+
+**Is your feature request related to a problem? Please describe.**
+A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]
+
+**Describe the solution you'd like**
+A clear and concise description of what you want to happen.
+
+**Describe alternatives you've considered**
+A clear and concise description of any alternative solutions or features you've considered.
+
+**Additional context**
+Add any other context or screenshots about the feature request here.

.github/ISSUE_TEMPLATE/🤔-questions-and-help.md

@@ -0,0 +1,10 @@
+---
+name: "\U0001F914 Questions and Help"
+about: Ask a usage or consultation question
+title: ''
+labels: ''
+assignees: ''
+
+---
+
+

.github/workflows/build-api-image.sh

@@ -1,61 +0,0 @@
-#!/usr/bin/env bash
-
-set -eo pipefail
-
-SHA=$(git rev-parse HEAD)
-REPO_NAME=langgenius/dify
-API_REPO_NAME="${REPO_NAME}-api"
-
-if [[ "${GITHUB_EVENT_NAME}" == "pull_request" ]]; then
-  REFSPEC=$(echo "${GITHUB_HEAD_REF}" | sed 's/[^a-zA-Z0-9]/-/g' | head -c 40)
-  PR_NUM=$(echo "${GITHUB_REF}" | sed 's:refs/pull/::' | sed 's:/merge::')
-  LATEST_TAG="pr-${PR_NUM}"
-  CACHE_FROM_TAG="latest"
-elif [[ "${GITHUB_EVENT_NAME}" == "release" ]]; then
-  REFSPEC=$(echo "${GITHUB_REF}" | sed 's:refs/tags/::' | head -c 40)
-  LATEST_TAG="${REFSPEC}"
-  CACHE_FROM_TAG="latest"
-else
-  REFSPEC=$(echo "${GITHUB_REF}" | sed 's:refs/heads/::' | sed 's/[^a-zA-Z0-9]/-/g' | head -c 40)
-  LATEST_TAG="${REFSPEC}"
-  CACHE_FROM_TAG="${REFSPEC}"
-fi
-
-if [[ "${REFSPEC}" == "main" ]]; then
-  LATEST_TAG="latest"
-  CACHE_FROM_TAG="latest"
-fi
-
-echo "Pulling cache image ${API_REPO_NAME}:${CACHE_FROM_TAG}"
-if docker pull "${API_REPO_NAME}:${CACHE_FROM_TAG}"; then
-  API_CACHE_FROM_SCRIPT="--cache-from ${API_REPO_NAME}:${CACHE_FROM_TAG}"
-else
-  echo "WARNING: Failed to pull ${API_REPO_NAME}:${CACHE_FROM_TAG}, disable build image cache."
-  API_CACHE_FROM_SCRIPT=""
-fi
-
-
-cat<<EOF
-  Rolling with tags:
-  - ${API_REPO_NAME}:${SHA}
-  - ${API_REPO_NAME}:${REFSPEC}
-  - ${API_REPO_NAME}:${LATEST_TAG}
-EOF
-
-#
-# Build image
-#
-cd api
-docker build \
-  ${API_CACHE_FROM_SCRIPT} \
-  --build-arg COMMIT_SHA=${SHA} \
-  -t "${API_REPO_NAME}:${SHA}" \
-  -t "${API_REPO_NAME}:${REFSPEC}" \
-  -t "${API_REPO_NAME}:${LATEST_TAG}" \
-  --label "sha=${SHA}" \
-  --label "built_at=$(date)" \
-  --label "build_actor=${GITHUB_ACTOR}" \
-  .
-
-# push
-docker push --all-tags "${API_REPO_NAME}"

.github/workflows/build-api-image.yml

@@ -5,18 +5,19 @@ on:
     branches:
       - 'main'
       - 'deploy/dev'
-  pull_request:
-    types: [synchronize, opened, reopened, ready_for_review]
+  release:
+    types: [published]
 
 jobs:
   build-and-push:
     runs-on: ubuntu-latest
     if: github.event.pull_request.draft == false
     steps:
-    - name: "Checkout ${{ github.ref }} ( ${{ github.sha }} )"
-      uses: actions/checkout@v2
-      with:
-        persist-credentials: false
+    - name: Set up QEMU
+      uses: docker/setup-qemu-action@v2
+
+    - name: Set up Docker Buildx
+      uses: docker/setup-buildx-action@v2
 
     - name: Login to Docker Hub
       uses: docker/login-action@v2
@@ -24,13 +25,29 @@ jobs:
         username: ${{ secrets.DOCKERHUB_USER }}
         password: ${{ secrets.DOCKERHUB_TOKEN }}
 
-    - name: Build and push Docker image
-      shell: bash
-      env:
-        DOCKERHUB_USER: ${{ secrets.DOCKERHUB_USER }}
-        DOCKERHUB_TOKEN: ${{ secrets.DOCKERHUB_TOKEN }}
-      run: |
-        /bin/bash .github/workflows/build-api-image.sh
+    - name: Extract metadata (tags, labels) for Docker
+      id: meta
+      uses: docker/metadata-action@v4
+      with:
+        images: langgenius/dify-api
+        tags: |
+          type=raw,value=latest,enable={{is_default_branch}}
+          type=ref,event=branch
+          type=sha,enable=true,priority=100,prefix=,suffix=,format=long
+          type=semver,pattern={{major}}.{{minor}}.{{patch}}
+          type=semver,pattern={{major}}.{{minor}}
+          type=semver,pattern={{major}}
+
+    - name: Build and push
+      uses: docker/build-push-action@v4
+      with:
+        context: "{{defaultContext}}:api"
+        platforms: linux/amd64,linux/arm64
+        build-args: |
+          COMMIT_SHA=${{ fromJSON(steps.meta.outputs.json).labels['org.opencontainers.image.revision'] }}
+        push: true
+        tags: ${{ steps.meta.outputs.tags }}
+        labels: ${{ steps.meta.outputs.labels }}
 
     - name: Deploy to server
       if: github.ref == 'refs/heads/deploy/dev'

.github/workflows/build-web-image.sh

@@ -1,60 +0,0 @@
-#!/usr/bin/env bash
-
-set -eo pipefail
-
-SHA=$(git rev-parse HEAD)
-REPO_NAME=langgenius/dify
-WEB_REPO_NAME="${REPO_NAME}-web"
-
-if [[ "${GITHUB_EVENT_NAME}" == "pull_request" ]]; then
-  REFSPEC=$(echo "${GITHUB_HEAD_REF}" | sed 's/[^a-zA-Z0-9]/-/g' | head -c 40)
-  PR_NUM=$(echo "${GITHUB_REF}" | sed 's:refs/pull/::' | sed 's:/merge::')
-  LATEST_TAG="pr-${PR_NUM}"
-  CACHE_FROM_TAG="latest"
-elif [[ "${GITHUB_EVENT_NAME}" == "release" ]]; then
-  REFSPEC=$(echo "${GITHUB_REF}" | sed 's:refs/tags/::' | head -c 40)
-  LATEST_TAG="${REFSPEC}"
-  CACHE_FROM_TAG="latest"
-else
-  REFSPEC=$(echo "${GITHUB_REF}" | sed 's:refs/heads/::' | sed 's/[^a-zA-Z0-9]/-/g' | head -c 40)
-  LATEST_TAG="${REFSPEC}"
-  CACHE_FROM_TAG="${REFSPEC}"
-fi
-
-if [[ "${REFSPEC}" == "main" ]]; then
-  LATEST_TAG="latest"
-  CACHE_FROM_TAG="latest"
-fi
-
-echo "Pulling cache image ${WEB_REPO_NAME}:${CACHE_FROM_TAG}"
-if docker pull "${WEB_REPO_NAME}:${CACHE_FROM_TAG}"; then
-  WEB_CACHE_FROM_SCRIPT="--cache-from ${WEB_REPO_NAME}:${CACHE_FROM_TAG}"
-else
-  echo "WARNING: Failed to pull ${WEB_REPO_NAME}:${CACHE_FROM_TAG}, disable build image cache."
-  WEB_CACHE_FROM_SCRIPT=""
-fi
-
-
-cat<<EOF
-  Rolling with tags:
-  - ${WEB_REPO_NAME}:${SHA}
-  - ${WEB_REPO_NAME}:${REFSPEC}
-  - ${WEB_REPO_NAME}:${LATEST_TAG}
-EOF
-
-#
-# Build image
-#
-cd web
-docker build \
-  ${WEB_CACHE_FROM_SCRIPT} \
-  --build-arg COMMIT_SHA=${SHA} \
-  -t "${WEB_REPO_NAME}:${SHA}" \
-  -t "${WEB_REPO_NAME}:${REFSPEC}" \
-  -t "${WEB_REPO_NAME}:${LATEST_TAG}" \
-  --label "sha=${SHA}" \
-  --label "built_at=$(date)" \
-  --label "build_actor=${GITHUB_ACTOR}" \
-  .
-
-docker push --all-tags "${WEB_REPO_NAME}"

.github/workflows/build-web-image.yml

@@ -5,18 +5,19 @@ on:
     branches:
       - 'main'
       - 'deploy/dev'
-  pull_request:
-    types: [synchronize, opened, reopened, ready_for_review]
+  release:
+    types: [published]
 
 jobs:
   build-and-push:
     runs-on: ubuntu-latest
     if: github.event.pull_request.draft == false
     steps:
-    - name: "Checkout ${{ github.ref }} ( ${{ github.sha }} )"
-      uses: actions/checkout@v2
-      with:
-        persist-credentials: false
+    - name: Set up QEMU
+      uses: docker/setup-qemu-action@v2
+
+    - name: Set up Docker Buildx
+      uses: docker/setup-buildx-action@v2
 
     - name: Login to Docker Hub
       uses: docker/login-action@v2
@@ -24,13 +25,29 @@ jobs:
         username: ${{ secrets.DOCKERHUB_USER }}
         password: ${{ secrets.DOCKERHUB_TOKEN }}
 
-    - name: Build and push Docker image
-      shell: bash
-      env:
-        DOCKERHUB_USER: ${{ secrets.DOCKERHUB_USER }}
-        DOCKERHUB_TOKEN: ${{ secrets.DOCKERHUB_TOKEN }}
-      run: |
-        /bin/bash .github/workflows/build-web-image.sh
+    - name: Extract metadata (tags, labels) for Docker
+      id: meta
+      uses: docker/metadata-action@v4
+      with:
+        images: langgenius/dify-web
+        tags: |
+          type=raw,value=latest,enable={{is_default_branch}}
+          type=ref,event=branch
+          type=sha,enable=true,priority=100,prefix=,suffix=,format=long
+          type=semver,pattern={{major}}.{{minor}}.{{patch}}
+          type=semver,pattern={{major}}.{{minor}}
+          type=semver,pattern={{major}}
+
+    - name: Build and push
+      uses: docker/build-push-action@v4
+      with:
+        context: "{{defaultContext}}:web"
+        platforms: linux/amd64,linux/arm64
+        build-args: |
+          COMMIT_SHA=${{ fromJSON(steps.meta.outputs.json).labels['org.opencontainers.image.revision'] }}
+        push: true
+        tags: ${{ steps.meta.outputs.tags }}
+        labels: ${{ steps.meta.outputs.labels }}
 
     - name: Deploy to server
       if: github.ref == 'refs/heads/deploy/dev'

.github/workflows/flake8.yml

@@ -1,19 +0,0 @@
-name: PEP8 Check
-on:
-  push:
-    branches: [main]
-  pull_request:
-    branches: [main]
-jobs:
-  pep8:
-    runs-on: ubuntu-latest
-    steps:
-      - uses: actions/checkout@v3
-      - name: Set up Python 3.10
-        uses: actions/setup-python@v4
-        with:
-          python-version: '3.10'
-      - name: Install flake8
-        run: pip install flake8
-      - name: Run flake8
-        run: flake8 --ignore=E501 .

.github/workflows/stale.yml

@@ -0,0 +1,29 @@
+# This workflow warns and then closes issues and PRs that have had no activity for a specified amount of time.
+#
+# You can adjust the behavior by modifying this file.
+# For more information, see:
+# https://github.com/actions/stale
+name: Mark stale issues and pull requests
+
+on:
+  schedule:
+  - cron: '0 3 * * *'
+
+jobs:
+  stale:
+
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write
+      pull-requests: write
+
+    steps:
+    - uses: actions/stale@v5
+      with:
+        days-before-issue-stale: 30
+        days-before-issue-close: 3
+        repo-token: ${{ secrets.GITHUB_TOKEN }}
+        stale-issue-message: "Close due to it's no longer active, if you have any questions, you can reopen it."
+        stale-pr-message: "Close due to it's no longer active, if you have any questions, you can reopen it."
+        stale-issue-label: 'no-issue-activity'
+        stale-pr-label: 'no-pr-activity'

.gitignore

@@ -130,7 +130,7 @@ dmypy.json
 .idea/'
 
 .DS_Store
-.vscode
+web/.vscode/settings.json
 
 # Intellij IDEA Files
 .idea/
@@ -139,7 +139,7 @@ dmypy.json
 api/.env
 api/storage/*
 
-docker/volumes/app/storage/privkeys/*
+docker/volumes/app/storage/*
 docker/volumes/db/data/*
 docker/volumes/redis/data/*
 docker/volumes/weaviate/*

CONTRIBUTING.md

@@ -22,14 +22,14 @@ To set up a working development environment, just fork the project git repositor
 
 ### Fork the repository
 
-you need to fork the [repository](https://github.com/langgenius/langgenius-gateway).
+you need to fork the [repository](https://github.com/langgenius/dify).
 
 ### Clone the repo
 
 Clone your GitHub forked repository:
 
 ```
-git clone git@github.com:<github_username>/langgenius-gateway.git
+git clone git@github.com:<github_username>/dify.git
 ```
 
 ### Install backend
@@ -54,3 +54,8 @@ Did you have an issue, like a merge conflict, or don't know how to open a pull r
 ## Community channels
 
 Stuck somewhere? Have any questions? Join the [Discord Community Server](https://discord.gg/AhzKf7dNgk). We are here to help!
+
+### i18n (Internationalization) Support
+
+We are looking for contributors to help with translations in other languages. If you are interested in helping, please join the [Discord Community Server](https://discord.gg/AhzKf7dNgk) and let us know.  
+Also check out the [Frontend i18n README]((web/i18n/README_EN.md)) for more information.

CONTRIBUTING_CN.md

@@ -51,3 +51,7 @@ git clone git@github.com:<github_username>/dify.git
 ## 社区渠道
 
 遇到困难了吗？有任何问题吗? 加入 [Discord Community Server](https://discord.gg/AhzKf7dNgk)，我们将为您提供帮助。
+
+### 多语言支持
+
+需要参与贡献翻译内容，请参阅[前端多语言翻译 README](web/i18n/README_CN.md)。

CONTRIBUTING_JA.md

@@ -0,0 +1,55 @@
+# コントリビュート
+
+[Dify](https://dify.ai) に興味を持ち、貢献したいと思うようになったことに感謝します！始める前に、
+[行動規範](https://github.com/langgenius/.github/blob/main/CODE_OF_CONDUCT.md)を読み、
+[既存の問題](https://github.com/langgenius/langgenius-gateway/issues)をチェックしてください。
+本ドキュメントは、[Dify](https://dify.ai) をビルドしてテストするための開発環境の構築方法を説明するものです。
+
+### 依存関係のインストール
+
+[Dify](https://dify.ai)をビルドするには、お使いのマシンに以下の依存関係をインストールし、設定する必要があります:
+
+- [Git](http://git-scm.com/)
+- [Docker](https://www.docker.com/)
+- [Docker Compose](https://docs.docker.com/compose/install/)
+- [Node.js v18.x (LTS)](http://nodejs.org)
+- [npm](https://www.npmjs.com/) バージョン 8.x.x もしくは [Yarn](https://yarnpkg.com/)
+- [Python](https://www.python.org/) バージョン 3.10.x
+
+## ローカル開発
+
+開発環境を構築するには、プロジェクトの git リポジトリをフォークし、適切なパッケージマネージャを使用してバックエンドとフロントエンドの依存関係をインストールし、docker-compose スタックを実行するように作成します。
+
+### リポジトリのフォーク
+
+[リポジトリ](https://github.com/langgenius/dify) をフォークする必要があります。
+
+### リポジトリのクローン
+
+GitHub でフォークしたリポジトリのクローンを作成する:
+
+```
+git clone git@github.com:<github_username>/dify.git
+```
+
+### バックエンドのインストール
+
+バックエンドアプリケーションのインストール方法については、[Backend README](api/README.md) を参照してください。
+
+### フロントエンドのインストール
+
+フロントエンドアプリケーションのインストール方法については、[Frontend README](web/README.md) を参照してください。
+
+### ブラウザで dify にアクセス
+
+[Dify](https://dify.ai) をローカル環境で見ることができるようになりました [http://localhost:3000](http://localhost:3000)。
+
+## プルリクエストの作成
+
+変更後、プルリクエスト (PR) をオープンしてください。プルリクエストを提出すると、Dify チーム/コミュニティの他の人があなたと一緒にそれをレビューします。
+
+マージコンフリクトなどの問題が発生したり、プルリクエストの開き方がわからなくなったりしませんでしたか？ [GitHub's pull request tutorial](https://docs.github.com/en/pull-requests/collaborating-with-pull-requests) で、マージコンフリクトやその他の問題を解決する方法をチェックしてみてください。あなたの PR がマージされると、[コントリビュータチャート](https://github.com/langgenius/langgenius-gateway/graphs/contributors)にコントリビュータとして誇らしげに掲載されます。
+
+## コミュニティチャンネル
+
+お困りですか？何か質問がありますか？ [Discord Community サーバ](https://discord.gg/AhzKf7dNgk)に参加してください。私たちがお手伝いします！

README.md

@@ -1,10 +1,12 @@
 ![](./images/describe-en.png)
 <p align="center">
   <a href="./README.md">English</a> |
-  <a href="./README_CN.md">简体中文</a>
+  <a href="./README_CN.md">简体中文</a> |
+  <a href="./README_JA.md">日本語</a> |
+  <a href="./README_ES.md">Español</a>
 </p>
 
-[Website](http://dify.ai) • [Docs](https://docs.dify.ai) • [Twitter](https://twitter.com/dify_ai)
+[Website](https://dify.ai) • [Docs](https://docs.dify.ai) • [Twitter](https://twitter.com/dify_ai) • [Discord](https://discord.gg/FngNHpbcY7)
 
 **Dify** is an easy-to-use LLMOps platform designed to empower more people to create sustainable, AI-native applications. With visual orchestration for various application types, Dify offers out-of-the-box, ready-to-use applications that can also serve as Backend-as-a-Service APIs. Unify your development process with one API for plugins and datasets integration, and streamline your operations using a single interface for prompt engineering, visual analytics, and continuous improvement.
 
@@ -21,7 +23,7 @@ Dify is compatible with Langchain, meaning we'll gradually support multiple LLMs
 
 ## Use Cloud Services
 
-Visit [Dify.ai](http://dify.ai)
+Visit [Dify.ai](https://dify.ai)
 
 ## Install the Community Edition
 
@@ -38,10 +40,15 @@ The easiest way to start the Dify server is to run our [docker-compose.yml](dock
 
 ```bash
 cd docker
-docker-compose up -d
+docker compose up -d
 ```
 
-After running, you can access the Dify console in your browser at [http://localhost](http://localhost) and start the initialization operation.
+After running, you can access the Dify dashboard in your browser at [http://localhost/install](http://localhost/install) and start the initialization installation process.
+
+### Helm Chart
+
+A big thanks to @BorisPolonsky for providing us with a [Helm Chart](https://helm.sh/) version, which allows Dify to be deployed on Kubernetes.
+You can go to https://github.com/BorisPolonsky/dify-helm for deployment information.
 
 ### Configuration
 
@@ -81,22 +88,42 @@ A: English and Chinese are currently supported, and you can contribute language
 
 [![Star History Chart](https://api.star-history.com/svg?repos=langgenius/dify&type=Date)](https://star-history.com/#langgenius/dify&Date)
 
+
+## Contributing 
+
+We welcome you to contribute to Dify to help make Dify better. We welcome contributions in various ways, submitting code, issues, new ideas, or sharing the interesting and useful AI applications you have created based on Dify. At the same time, we also welcome you to share Dify at different events, conferences, and social media.
+
+### Submit a Pull Request 
+
+To ensure proper review, all code contributions, including from contributors with direct commit access, must be submitted as PR requests and approved by core developers before merging branches. 
+We welcome PRs from everyone! If you're willing to help out, you can learn more about how to contribute code to the project in the [Contribution Guide](CONTRIBUTING.md).  
+
+### Submit issues or ideas  
+
+You can submit your issues or ideas by adding issues to the Dify repository. If you encounter issues, please describe the steps you took to encounter the issue as much as possible so we can better discover it. If you have any new ideas for our product, we also welcome your feedback. Please share your insights as much as possible so we can get more feedback and further discussion in the community.  
+
+### Share your applications
+
+We encourage all community members to share their AI applications built on Dify, which can be applied to different scenarios or different users. This will provide powerful inspiration for people who want to create AI capabilities! You can share your experience by [submitting an issue in the Dify-user-case repository](https://github.com/langgenius/dify-user-case/issues).  
+
+### Share Dify with others
+
+We encourage community contributors to actively demonstrate different aspects of using Dify. You can talk or share any feature of using Dify at  meetups and conferences, blogs or social media. We believe your unique sharing will be of great help to others!  Mention @Dify.AI on Twitter and/or communicate on [Discord](https://discord.gg/FngNHpbcY7) so we can give pointers and tips and help you spread the word by promoting your content on the different Dify communication channels.
+
+### Help others 
+You can also help people in need of help on Discord, GitHub issues or other social platforms, guide others to solve problems encountered during use and share usage experiences. This is also a great contribution! If you want to become a maintainer of the Dify community, please contact the official team via [Discord](https://discord.gg/FngNHpbcY7) or email us at support@dify.ai. 
+
+
 ## Contact Us
 
 If you have any questions, suggestions, or partnership inquiries, feel free to contact us through the following channels:
 
 - Submit an Issue or PR on our GitHub Repo
-- Join the discussion in our [Discord](https://discord.gg/AhzKf7dNgk) Community
+- Join the discussion in our [Discord](https://discord.gg/FngNHpbcY7) Community
 - Send an email to hello@dify.ai
 
 We're eager to assist you and together create more fun and useful AI applications!
 
-## Contributing
-
-To ensure proper review, all code contributions - including those from contributors with direct commit access - must be submitted via pull requests and approved by the core development team prior to being merged.
-
-We welcome all pull requests! If you'd like to help, check out the [Contribution Guide](CONTRIBUTING.md) for more information on how to get started.
-
 ## Security
 
 To protect your privacy, please avoid posting security issues on GitHub. Instead, send your questions to security@dify.ai and we will provide you with a more detailed answer.

README_CN.md

@@ -1,11 +1,13 @@
 ![](./images/describe-cn.jpg)
 <p align="center">
   <a href="./README.md">English</a> |
-  <a href="./README_CN.md">简体中文</a>
+  <a href="./README_CN.md">简体中文</a> |
+  <a href="./README_JA.md">日本語</a> |
+  <a href="./README_ES.md">Español</a>
 </p>
 
 
-[官方网站](http://dify.ai) • [文档](https://docs.dify.ai/v/zh-hans) • [Twitter](https://twitter.com/dify_ai)
+[官方网站](https://dify.ai) • [文档](https://docs.dify.ai/v/zh-hans) • [Twitter](https://twitter.com/dify_ai) •  [Discord](https://discord.gg/FngNHpbcY7)
 
 **Dify** 是一个易用的 LLMOps 平台，旨在让更多人可以创建可持续运营的原生 AI 应用。Dify 提供多种类型应用的可视化编排，应用可开箱即用，也能以“后端即服务”的 API 提供服务。
 
@@ -23,7 +25,7 @@ Dify 兼容 Langchain，这意味着我们将逐步支持多种 LLMs ，目前
 
 ## 使用云服务
 
-访问 [Dify.ai](http://cloud.dify.ai)
+访问 [Dify.ai](https://cloud.dify.ai)
 
 ## 安装社区版
 
@@ -40,10 +42,15 @@ Dify 兼容 Langchain，这意味着我们将逐步支持多种 LLMs ，目前
 
 ```bash
 cd docker
-docker-compose up -d
+docker compose up -d
 ```
 
-运行后，可以在浏览器上访问 [http://localhost](http://localhost) 进入 Dify 控制台，并开始初始化操作。
+运行后，可以在浏览器上访问 [http://localhost/install](http://localhost/install) 进入 Dify 控制台并开始初始化安装操作。
+
+### Helm Chart
+
+非常感谢 @BorisPolonsky 为我们提供了一个 [Helm Chart](https://helm.sh/) 版本，可以在 Kubernetes 上部署 Dify。
+您可以前往 https://github.com/BorisPolonsky/dify-helm 来获取部署信息。
 
 ### 配置
 
@@ -82,20 +89,37 @@ A: 现已支持英文与中文，你可以为我们贡献语言包。
 
 [![Star History Chart](https://api.star-history.com/svg?repos=langgenius/dify&type=Date)](https://star-history.com/#langgenius/dify&Date)
 
+
+## 贡献
+
+我们欢迎你为 Dify 作出贡献帮助 Dify 变得更好。我们欢迎各种方式的贡献，提交代码、问题、新想法、或者分享你基于 Dify 创建出的各种有趣有用的 AI 应用。同时，我们也欢迎你在不同的活动、研讨会、社交媒体上分享 Dify。
+
+### 贡献代码
+为了确保正确审查，所有代码贡献 - 包括来自具有直接提交更改权限的贡献者 - 都必须提交 PR 请求并在合并分支之前得到核心开发人员的批准。
+
+我们欢迎所有人提交 PR！如果您愿意提供帮助，可以在 [贡献指南](CONTRIBUTING_CN.md) 中了解有关如何为项目做出代码贡献的更多信息。
+
+### 提交问题或想法
+你可以通过 Dify 代码仓库新增 issues 来提交你的问题或想法。如遇到问题，请尽可能描述你遇到问题的操作步骤，以便我们更好地发现它。如果你对我们的产品有任何新想法，也欢迎向我们反馈，请尽可能多地分享你的见解，以便我们在社区中获得更多反馈和进一步讨论。
+
+### 分享你的应用
+我们鼓励所有社区成员分享他们基于 Dify 创造出的 AI 应用，它们可以是应用于不同情景或不同用户，这将有助于为希望基于 AI 能力创造的人们提供强大灵感！你可以通过 [Dify-user-case 仓库项目提交 issue](https://github.com/langgenius/dify-user-case) 来分享你的应用案例。
+
+### 向别人分享 Dify
+我们鼓励社区贡献者们积极展示你使用 Dify 的不同角度。你可以通过线下研讨会、博客或社交媒体上谈论或分享你使用 Dify 的任意功能，相信你独特的使用分享会给别人带来非常大的帮助！如果你需要任何指导帮助，欢迎联系我们 support@dify.ai ,你也可以在 twitter @Dify.AI 或在 [Discord 社区](https://discord.gg/FngNHpbcY7)交流来帮助你传播信息。
+
+### 帮助别人
+你还可以在 Discord、GitHub issues或其他社交平台上帮助需要帮助的人，指导别人解决使用过程中遇到的问题和分享使用经验。这也是个非常了不起的贡献！如果你希望成为 Dify 社区的维护者，请通过[Discord 社区](https://discord.gg/FngNHpbcY7) 联系官方团队或邮件联系我们 support@dify.ai.
+
+
 ## 联系我们
 
 如果您有任何问题、建议或合作意向，欢迎通过以下方式联系我们：
 
 - 在我们的 [GitHub Repo](https://github.com/langgenius/dify) 上提交 Issue 或 PR
-- 在我们的 [Discord 社区](https://discord.gg/AhzKf7dNgk) 上加入讨论
+- 在我们的 [Discord 社区](https://discord.gg/FngNHpbcY7) 上加入讨论
 - 发送邮件至 hello@dify.ai
 
-## 贡献代码
-
-为了确保正确审查，所有代码贡献 - 包括来自具有直接提交更改权限的贡献者 - 都必须提交 PR 请求并在合并分支之前得到核心开发人员的批准。
-
-我们欢迎所有人提交 PR！如果您愿意提供帮助，可以在 [贡献指南](CONTRIBUTING_CN.md) 中了解有关如何为项目做出贡献的更多信息。
-
 ## 安全
 
 为了保护您的隐私，请避免在 GitHub 上发布安全问题。发送问题至 security@dify.ai，我们将为您做更细致的解答。

README_ES.md

@@ -0,0 +1,124 @@
+![](./images/describe-en.png)
+<p align="center">
+  <a href="./README.md">English</a> |
+  <a href="./README_CN.md">简体中文</a> |
+  <a href="./README_JA.md">日本語</a> |
+  <a href="./README_ES.md">Español</a>
+</p>
+
+[Sitio web](https://dify.ai) • [Documentación](https://docs.dify.ai) • [Twitter](https://twitter.com/dify_ai) • [Discord](https://discord.gg/FngNHpbcY7)
+
+**Dify** es una plataforma LLMOps fácil de usar diseñada para capacitar a más personas para que creen aplicaciones sostenibles basadas en IA. Con orquestación visual para varios tipos de aplicaciones, Dify ofrece aplicaciones listas para usar que también pueden funcionar como APIs de Backend-as-a-Service. Unifica tu proceso de desarrollo con una API para la integración de complementos y conjuntos de datos, y agiliza tus operaciones utilizando una interfaz única para la ingeniería de indicaciones, análisis visual y mejora continua.
+
+Las aplicaciones creadas con Dify incluyen:
+
+- Sitios web listos para usar que admiten el modo de formulario y el modo de conversación por chat.
+- Una API única que abarca capacidades de complementos, mejora de contexto y más, lo que te ahorra esfuerzo de programación en el backend.
+- Análisis visual de datos, revisión de registros y anotación para aplicaciones.
+
+Dify es compatible con Langchain, lo que significa que gradualmente admitiremos múltiples LLMs, actualmente compatibles con:
+
+- GPT 3 (text-davinci-003)
+- GPT 3.5 Turbo (ChatGPT)
+- GPT-4
+
+## Usar servicios en la nube
+
+Visita [Dify.ai](https://dify.ai)
+
+## Instalar la Edición Comunitaria
+
+### Requisitos del sistema
+
+Antes de instalar Dify, asegúrate de que tu máquina cumple con los siguientes requisitos mínimos del sistema:
+
+- CPU >= 1 Core
+- RAM >= 4GB
+
+### Inicio rápido
+
+La forma más sencilla de iniciar el servidor de Dify es ejecutar nuestro archivo [docker-compose.yml](docker/docker-compose.yaml). Antes de ejecutar el comando de instalación, asegúrate de que [Docker](https://docs.docker.com/get-docker/) y [Docker Compose](https://docs.docker.com/compose/install/) estén instalados en tu máquina:
+
+```bash
+cd docker
+docker compose up -d
+```
+
+Después de ejecutarlo, puedes acceder al panel de control de Dify en tu navegador desde [http://localhost/install](http://localhost/install) y comenzar el proceso de instalación de inicialización.
+
+### Helm Chart
+
+Un gran agradecimiento a @BorisPolonsky por proporcionarnos una versión de [Helm Chart](https://helm.sh/), que permite desplegar Dify en Kubernetes.
+Puede ir a https://github.com/BorisPolonsky/dify-helm para obtener información de despliegue.
+
+### Configuración
+
+Si necesitas personalizar la configuración, consulta los comentarios en nuestro archivo [docker-compose.yml](docker/docker-compose.yaml) y configura manualmente la configuración del entorno. Después de realizar los cambios, ejecuta nuevamente 'docker-compose up -d'.
+
+## Hoja de ruta
+
+Funciones en desarrollo:
+
+- **Conjuntos de datos**, admitiendo más conjuntos de datos, por ejemplo, sincronización de contenido desde Notion o páginas web.
+Admitiremos más conjuntos de datos, incluidos texto, páginas web e incluso contenido de Notion. Los usuarios pueden construir aplicaciones de IA basadas en sus propias fuentes de datos
+- **Complementos**, introduciendo complementos estándar de ChatGPT para aplicaciones, o utilizando complementos producidos por Dify.
+Lanzaremos complementos que cumplan con el estándar de ChatGPT, o nuestros propios complementos de Dify para habilitar más capacidades en las aplicaciones.
+- **Modelos de código abierto**, por ejemplo, adoptar Llama como proveedor de modelos o para un ajuste adicional.
+Trabajaremos con excelentes modelos de código abierto como Llama, proporcionándolos como opciones de modelos en nuestra plataforma o utilizándolos para un ajuste adicional.
+
+## Preguntas y respuestas
+
+**P: ¿Qué puedo hacer con Dify?**
+
+R: Dify es una herramienta de desarrollo y operaciones de LLM, simple pero poderosa. Puedes usarla para construir aplicaciones de calidad comercial y asistentes personales. Si deseas desarrollar tus propias aplicaciones, LangDifyGenius puede ahorrarte trabajo en el backend al integrar con OpenAI y ofrecer capacidades de operaciones visuales, lo que te permite mejorar y entrenar continuamente tu modelo GPT.
+
+**P: ¿Cómo uso Dify para "entrenar" mi propio modelo?**
+
+R: Una aplicación valiosa consta de Ingeniería de indicaciones, mejora de contexto y ajuste fino. Hemos creado un enfoque de programación híbrida que combina las indicaciones con lenguajes de programación (similar a un motor de plantillas), lo que facilita la incorporación de texto largo o la captura de subtítulos de un video de YouTube ingresado por el usuario, todo lo cual se enviará como contexto para que los LLM lo procesen. Damos gran importancia a la operabilidad de la aplicación, con los datos generados por los usuarios durante el uso de la aplicación disponibles para análisis, anotación y entrenamiento continuo. Sin las herramientas adecuadas, estos pasos pueden llevar mucho tiempo.
+
+**P: ¿Qué necesito preparar si quiero crear mi propia aplicación?**
+
+R: Suponemos que ya tienes una clave de API de OpenAI; si no la tienes, por favor regístrate. ¡Si ya tienes contenido que pueda servir como contexto de entrenamiento, eso es genial!
+
+**P: ¿Qué idiomas de interfaz están disponibles?**
+
+R: Actualmente se admiten inglés y chino, y puedes contribuir con paquetes de idiomas.
+
+## Historial de estrellas
+
+[![Gráfico de historial de estrellas](https://api.star-history.com/svg?repos=langgenius/dify&type=Date)](https://star-history.com/#langgenius/dify&Date)
+
+## Contáctanos
+
+Si tienes alguna pregunta, sugerencia o consulta sobre asociación, no dudes en contactarnos a través de los siguientes canales:
+
+- Presentar un problema o una solicitud de extracción en nuestro repositorio de GitHub.
+- Únete a la discusión en nuestra comunidad de [Discord](https://discord.gg/FngNHpbcY7).
+- Envía un correo electrónico a hello@dify.ai.
+
+¡Estamos ansiosos por ayudarte y crear juntos aplicaciones de IA más divertidas y útiles!
+
+## Contribuciones
+
+Para garantizar una revisión adecuada, todas las contribuciones de código, incluidas las de los colaboradores con acceso directo a los compromisos, deben enviarse mediante solicitudes de extracción y ser aprobadas por el equipo principal de
+
+ desarrollo antes de fusionarse.
+
+¡Agradecemos todas las solicitudes de extracción! Si deseas ayudar, consulta la [Guía de Contribución](CONTRIBUTING.md) para obtener más información sobre cómo comenzar.
+
+## Seguridad
+
+Para proteger tu privacidad, evita publicar problemas de seguridad en GitHub. En su lugar, envía tus preguntas a security@dify.ai y te proporcionaremos una respuesta más detallada.
+
+## Citación
+
+Este software utiliza el siguiente software de código abierto:
+
+- Chase, H. (2022). LangChain [Software de computadora]. https://github.com/hwchase17/langchain
+- Liu, J. (2022). LlamaIndex [Software de computadora]. doi: 10.5281/zenodo.1234.
+
+Para obtener más información, consulta el sitio web oficial o el texto de la licencia del software correspondiente.
+
+## Licencia
+
+Este repositorio está disponible bajo la [Licencia de código abierto de Dify](LICENSE).

README_JA.md

@@ -0,0 +1,123 @@
+![](./images/describe-en.png)
+<p align="center">
+  <a href="./README.md">English</a> |
+  <a href="./README_CN.md">简体中文</a> |
+  <a href="./README_JA.md">日本語</a> |
+  <a href="./README_ES.md">Español</a>
+</p>
+
+[Web サイト](https://dify.ai) • [ドキュメント](https://docs.dify.ai) • [Twitter](https://twitter.com/dify_ai) • [Discord](https://discord.gg/FngNHpbcY7)
+
+
+**Dify** は、より多くの人々が持続可能な AI ネイティブアプリケーションを作成できるように設計された、使いやすい LLMOps プラットフォームです。様々なアプリケーションタイプに対応したビジュアルオーケストレーションにより Dify は Backend-as-a-Service API としても機能する、すぐに使えるアプリケーションを提供します。プラグインやデータセットを統合するための1つの API で開発プロセスを統一し、プロンプトエンジニアリング、ビジュアル分析、継続的な改善のための1つのインターフェイスを使って業務を合理化します。
+
+Difyで作成したアプリケーションは以下の通りです:
+
+フォームモードとチャット会話モードをサポートする、すぐに使える Web サイト
+プラグイン機能、コンテキストの強化などを網羅する単一の API により、バックエンドのコーディングの手間を省きます。
+アプリケーションの視覚的なデータ分析、ログレビュー、アノテーションが可能です。
+Dify は LangChain と互換性があり、複数の LLM を徐々にサポートします:
+
+- GPT 3 (text-davinci-003)
+- GPT 3.5 Turbo(ChatGPT)
+- GPT-4
+
+## クラウドサービスの利用
+
+[Dify.ai](https://dify.ai) をご覧ください
+
+## Community Edition のインストール
+
+### システム要件
+
+Dify をインストールする前に、お使いのマシンが以下の最低システム要件を満たしていることを確認してください:
+
+- CPU >= 1 Core
+- RAM >= 4GB
+
+### クイックスタート
+
+Dify サーバーを起動する最も簡単な方法は、[docker-compose.yml](docker/docker-compose.yaml) ファイルを実行することです。インストールコマンドを実行する前に、[Docker](https://docs.docker.com/get-docker/) と [Docker Compose](https://docs.docker.com/compose/install/) がお使いのマシンにインストールされていることを確認してください:
+
+```bash
+cd docker
+docker compose up -d
+```
+
+実行後、ブラウザで [http://localhost/install](http://localhost/install) にアクセスし、初期化インストール作業を開始することができます。
+
+### Helm Chart
+
+@BorisPolonsky に大感謝します。彼は Dify を Kubernetes 上にデプロイするための [Helm Chart](https://helm.sh/) バージョンを提供してくれました。
+デプロイ情報については、https://github.com/BorisPolonsky/dify-helm をご覧ください。
+
+### 構成
+
+カスタマイズが必要な場合は、[docker-compose.yml](docker/docker-compose.yaml) ファイルのコメントを参照し、手動で環境設定をお願いします。変更後、再度 'docker-compose up -d' を実行してください。
+
+## ロードマップ
+
+開発中の機能:
+
+- **データセット**, Notionやウェブページからのコンテンツ同期など、より多くのデータセットをサポートします
+テキスト、ウェブページ、さらには Notion コンテンツなど、より多くのデータセットをサポートする予定です。ユーザーは、自分のデータソースをもとに AI アプリケーションを構築することができます。
+- **プラグイン**, アプリケーションに ChatGPT プラグイン標準のプラグインを導入する、または Dify 制作のプラグインを利用する
+今後、ChatGPT 規格に準拠したプラグインや、ディファイ独自のプラグインを公開し、より多くの機能をアプリケーションで実現できるようにします。
+- **オープンソースモデル**, 例えばモデルプロバイダーとして Llama を採用したり、さらにファインチューニングを行う
+Llama のような優れたオープンソースモデルを、私たちのプラットフォームのモデルオプションとして提供したり、さらなる微調整のために使用したりすることで、協力していきます。
+
+
+## Q&A
+
+**Q: Dify で何ができるのか？**
+
+A: Dify はシンプルでパワフルな LLM 開発・運用ツールです。商用グレードのアプリケーション、パーソナルアシスタントを構築するために使用することができます。独自のアプリケーションを開発したい場合、LangDifyGenius は OpenAI と統合する際のバックエンド作業を省き、視覚的な操作機能を提供し、GPT モデルを継続的に改善・訓練することが可能です。
+
+**Q: Dify を使って、自分のモデルを「トレーニング」するにはどうすればいいのでしょうか？**
+
+A: プロンプトエンジニアリング、コンテキスト拡張、ファインチューニングからなる価値あるアプリケーションです。プロンプトとプログラミング言語を組み合わせたハイブリッドプログラミングアプローチ（テンプレートエンジンのようなもの）で、長文の埋め込みやユーザー入力の YouTube 動画からの字幕取り込みなどを簡単に実現し、これらはすべて LLM が処理するコンテキストとして提出される予定です。また、アプリケーションの操作性を重視し、ユーザーがアプリケーションを使用する際に生成したデータを分析、アノテーション、継続的なトレーニングに利用できるようにしました。適切なツールがなければ、これらのステップに時間がかかることがあります。
+
+**Q: 自分でアプリケーションを作りたい場合、何を準備すればよいですか？**
+
+A: すでに OpenAI API Key をお持ちだと思いますが、お持ちでない場合はご登録ください。もし、すでにトレーニングのコンテキストとなるコンテンツをお持ちでしたら、それは素晴らしいことです！
+
+**Q: インターフェイスにどの言語が使えますか？**
+
+A: 現在、英語と中国語に対応しており、言語パックを寄贈することも可能です。
+
+## Star ヒストリー
+
+[![Star History Chart](https://api.star-history.com/svg?repos=langgenius/dify&type=Date)](https://star-history.com/#langgenius/dify&Date)
+
+## お問合せ
+
+ご質問、ご提案、パートナーシップに関するお問い合わせは、以下のチャンネルからお気軽にご連絡ください:
+
+- GitHub Repo で Issue や PR を提出する
+- [Discord](https://discord.gg/FngNHpbcY7) コミュニティで議論に参加する。
+- hello@dify.ai にメールを送信します
+
+私たちは、皆様のお手伝いをさせていただき、より楽しく、より便利な AI アプリケーションを一緒に作っていきたいと思っています！
+
+## コントリビュート
+
+適切なレビューを行うため、コミットへの直接アクセスが可能なコントリビュータを含むすべてのコードコントリビュータは、プルリクエストで提出し、マージされる前にコア開発チームによって承認される必要があります。
+
+私たちはすべてのプルリクエストを歓迎します！協力したい方は、[コントリビューションガイド](CONTRIBUTING.md) をチェックしてみてください。
+
+## セキュリティ
+
+プライバシー保護のため、GitHub へのセキュリティ問題の投稿は避けてください。代わりに、あなたの質問を security@dify.ai に送ってください。より詳細な回答を提供します。
+
+## 引用
+
+本ソフトウェアは、以下のオープンソースソフトウェアを使用しています:
+
+- Chase, H. (2022). LangChain [Computer software]. https://github.com/hwchase17/langchain
+- Liu, J. (2022). LlamaIndex [Computer software]. doi: 10.5281/zenodo.1234.
+
+詳しくは、各ソフトウェアの公式サイトまたはライセンス文をご参照ください。
+
+## ライセンス
+
+このリポジトリは、[Dify Open Source License](LICENSE) のもとで利用できます。

api/.env.example

@@ -14,7 +14,7 @@ CONSOLE_URL=http://127.0.0.1:5001
 API_URL=http://127.0.0.1:5001
 
 # Web APP base URL
-APP_URL=http://127.0.0.1:5001
+APP_URL=http://127.0.0.1:3000
 
 # celery configuration
 CELERY_BROKER_URL=redis://:difyai123456@localhost:6379/1
@@ -22,6 +22,7 @@ CELERY_BROKER_URL=redis://:difyai123456@localhost:6379/1
 # redis configuration
 REDIS_HOST=localhost
 REDIS_PORT=6379
+REDIS_USERNAME: ''
 REDIS_PASSWORD=difyai123456
 REDIS_DB=0
 
@@ -72,6 +73,7 @@ VECTOR_STORE=weaviate
 WEAVIATE_ENDPOINT=http://localhost:8080
 WEAVIATE_API_KEY=WVF5YThaHlkYwhGUSmCRgsX3tD5ngdN8pkih
 WEAVIATE_GRPC_ENABLED=false
+WEAVIATE_BATCH_SIZE=100
 
 # Qdrant configuration, use `path:` prefix for local mode or `https://your-qdrant-cluster-url.qdrant.io` for remote mode
 QDRANT_URL=path:storage/qdrant
@@ -83,3 +85,9 @@ SENTRY_DSN=
 # DEBUG
 DEBUG=false
 SQLALCHEMY_ECHO=false
+
+# Notion import configuration, support public and internal
+NOTION_INTEGRATION_TYPE=public
+NOTION_CLIENT_SECRET=you-client-secret
+NOTION_CLIENT_ID=you-client-id
+NOTION_INTERNAL_SECRET=you-internal-secret

api/README.md

@@ -17,6 +17,11 @@
    ```bash
    openssl rand -base64 42
    ```
+3.5 If you use annaconda, create a new environment and activate it
+   ```bash
+   conda create --name dify python=3.10
+   conda activate dify
+   ```
 4. Install dependencies
    ```bash
    pip install -r requirements.txt
@@ -33,3 +38,4 @@
    flask run --host 0.0.0.0 --port=5001 --debug
    ```
 7. Setup your application by visiting http://localhost:5001/console/api/setup or other apis...
+8. If you need to debug local async processing, you can run `celery -A app.celery worker`, celery can do dataset importing and other async tasks.

api/app.py

@@ -1,5 +1,7 @@
 # -*- coding:utf-8 -*-
 import os
+from datetime import datetime
+
 if not os.environ.get("DEBUG") or os.environ.get("DEBUG").lower() != 'true':
     from gevent import monkey
     monkey.patch_all()
@@ -18,7 +20,7 @@ from extensions.ext_database import db
 from extensions.ext_login import login_manager
 
 # DO NOT REMOVE BELOW
-from models import model, account, dataset, web, task
+from models import model, account, dataset, web, task, source
 from events import event_handlers
 # DO NOT REMOVE ABOVE
 
@@ -122,6 +124,9 @@ def load_user(user_id):
                     account.current_tenant_id = tenant_account_join.tenant_id
                     session['workspace_id'] = account.current_tenant_id
 
+            account.last_active_at = datetime.utcnow()
+            db.session.commit()
+
             # Log in the user with the updated user_id
             flask_login.login_user(account, remember=True)
 

api/commands.py

@@ -1,18 +1,21 @@
 import datetime
-import json
 import random
 import string
 
 import click
+from flask import current_app
 
 from libs.password import password_pattern, valid_password, hash_password
 from libs.helper import email as email_validate
 from extensions.ext_database import db
-from models.account import InvitationCode
-from models.model import Account, AppModelConfig, ApiToken, Site, App, RecommendedApp
+from libs.rsa import generate_key_pair
+from models.account import InvitationCode, Tenant
+from models.model import Account
 import secrets
 import base64
 
+from models.provider import Provider
+
 
 @click.command('reset-password', help='Reset the account password.')
 @click.option('--email', prompt=True, help='The email address of the account whose password you need to reset')
@@ -74,6 +77,31 @@ def reset_email(email, new_email, email_confirm):
     click.echo(click.style('Congratulations!, email has been reset.', fg='green'))
 
 
+@click.command('reset-encrypt-key-pair', help='Reset the asymmetric key pair of workspace for encrypt LLM credentials. '
+                                              'After the reset, all LLM credentials will become invalid, '
+                                              'requiring re-entry.'
+                                              'Only support SELF_HOSTED mode.')
+@click.confirmation_option(prompt=click.style('Are you sure you want to reset encrypt key pair?'
+                                              ' this operation cannot be rolled back!', fg='red'))
+def reset_encrypt_key_pair():
+    if current_app.config['EDITION'] != 'SELF_HOSTED':
+        click.echo(click.style('Sorry, only support SELF_HOSTED mode.', fg='red'))
+        return
+
+    tenant = db.session.query(Tenant).first()
+    if not tenant:
+        click.echo(click.style('Sorry, no workspace found. Please enter /install to initialize.', fg='red'))
+        return
+
+    tenant.encrypt_public_key = generate_key_pair(tenant.id)
+
+    db.session.query(Provider).filter(Provider.provider_type == 'custom').delete()
+    db.session.commit()
+
+    click.echo(click.style('Congratulations! '
+                           'the asymmetric key pair of workspace {} has been reset.'.format(tenant.id), fg='green'))
+
+
 @click.command('generate-invitation-codes', help='Generate invitation codes.')
 @click.option('--batch', help='The batch of invitation codes.')
 @click.option('--count', prompt=True, help='Invitation codes count.')
@@ -131,30 +159,8 @@ def generate_upper_string():
     return result
 
 
-@click.command('gen-recommended-apps', help='Number of records to generate')
-def generate_recommended_apps():
-    print('Generating recommended app data...')
-    apps = App.query.all()
-    for app in apps:
-        recommended_app = RecommendedApp(
-            app_id=app.id,
-            description={
-                'en': 'Description for ' + app.name,
-                'zh': '描述 ' + app.name
-            },
-            copyright='Copyright ' + str(random.randint(1990, 2020)),
-            privacy_policy='https://privacypolicy.example.com',
-            category=random.choice(['Games', 'News', 'Music', 'Sports']),
-            position=random.randint(1, 100),
-            install_count=random.randint(100, 100000)
-        )
-        db.session.add(recommended_app)
-    db.session.commit()
-    print('Done!')
-
-
 def register_commands(app):
     app.cli.add_command(reset_password)
     app.cli.add_command(reset_email)
     app.cli.add_command(generate_invitation_codes)
-    app.cli.add_command(generate_recommended_apps)
+    app.cli.add_command(reset_encrypt_key_pair)

api/config.py

@@ -21,9 +21,11 @@ DEFAULTS = {
     'REDIS_HOST': 'localhost',
     'REDIS_PORT': '6379',
     'REDIS_DB': '0',
+    'REDIS_USE_SSL': 'False',
     'SESSION_REDIS_HOST': 'localhost',
     'SESSION_REDIS_PORT': '6379',
     'SESSION_REDIS_DB': '2',
+    'SESSION_REDIS_USE_SSL': 'False',
     'OAUTH_REDIRECT_PATH': '/console/api/oauth/authorize',
     'OAUTH_REDIRECT_INDEX_PATH': '/',
     'CONSOLE_URL': 'https://cloud.dify.ai',
@@ -41,9 +43,12 @@ DEFAULTS = {
     'SENTRY_TRACES_SAMPLE_RATE': 1.0,
     'SENTRY_PROFILES_SAMPLE_RATE': 1.0,
     'WEAVIATE_GRPC_ENABLED': 'True',
+    'WEAVIATE_BATCH_SIZE': 100,
     'CELERY_BACKEND': 'database',
     'PDF_PREVIEW': 'True',
     'LOG_LEVEL': 'INFO',
+    'DISABLE_PROVIDER_CONFIG_VALIDATION': 'False',
+    'DEFAULT_LLM_PROVIDER': 'openai'
 }
 
 
@@ -74,7 +79,7 @@ class Config:
         self.CONSOLE_URL = get_env('CONSOLE_URL')
         self.API_URL = get_env('API_URL')
         self.APP_URL = get_env('APP_URL')
-        self.CURRENT_VERSION = "0.2.0"
+        self.CURRENT_VERSION = "0.3.4"
         self.COMMIT_SHA = get_env('COMMIT_SHA')
         self.EDITION = "SELF_HOSTED"
         self.DEPLOY_ENV = get_env('DEPLOY_ENV')
@@ -105,14 +110,18 @@ class Config:
         # redis settings
         self.REDIS_HOST = get_env('REDIS_HOST')
         self.REDIS_PORT = get_env('REDIS_PORT')
+        self.REDIS_USERNAME = get_env('REDIS_USERNAME')
         self.REDIS_PASSWORD = get_env('REDIS_PASSWORD')
         self.REDIS_DB = get_env('REDIS_DB')
+        self.REDIS_USE_SSL = get_bool_env('REDIS_USE_SSL')
 
         # session redis settings
         self.SESSION_REDIS_HOST = get_env('SESSION_REDIS_HOST')
         self.SESSION_REDIS_PORT = get_env('SESSION_REDIS_PORT')
+        self.SESSION_REDIS_USERNAME = get_env('SESSION_REDIS_USERNAME')
         self.SESSION_REDIS_PASSWORD = get_env('SESSION_REDIS_PASSWORD')
         self.SESSION_REDIS_DB = get_env('SESSION_REDIS_DB')
+        self.SESSION_REDIS_USE_SSL = get_bool_env('SESSION_REDIS_USE_SSL')
 
         # storage settings
         self.STORAGE_TYPE = get_env('STORAGE_TYPE')
@@ -130,6 +139,7 @@ class Config:
         self.WEAVIATE_ENDPOINT = get_env('WEAVIATE_ENDPOINT')
         self.WEAVIATE_API_KEY = get_env('WEAVIATE_API_KEY')
         self.WEAVIATE_GRPC_ENABLED = get_bool_env('WEAVIATE_GRPC_ENABLED')
+        self.WEAVIATE_BATCH_SIZE = int(get_env('WEAVIATE_BATCH_SIZE'))
 
         # qdrant settings
         self.QDRANT_URL = get_env('QDRANT_URL')
@@ -165,10 +175,24 @@ class Config:
         self.CELERY_BACKEND = get_env('CELERY_BACKEND')
         self.CELERY_RESULT_BACKEND = 'db+{}'.format(self.SQLALCHEMY_DATABASE_URI) \
             if self.CELERY_BACKEND == 'database' else self.CELERY_BROKER_URL
+        self.BROKER_USE_SSL = self.CELERY_BROKER_URL.startswith('rediss://')
 
         # hosted provider credentials
         self.OPENAI_API_KEY = get_env('OPENAI_API_KEY')
 
+        # By default it is False
+        # You could disable it for compatibility with certain OpenAPI providers
+        self.DISABLE_PROVIDER_CONFIG_VALIDATION = get_bool_env('DISABLE_PROVIDER_CONFIG_VALIDATION')
+
+        # For temp use only
+        # set default LLM provider, default is 'openai', support `azure_openai`
+        self.DEFAULT_LLM_PROVIDER = get_env('DEFAULT_LLM_PROVIDER')
+        # notion import setting
+        self.NOTION_CLIENT_ID = get_env('NOTION_CLIENT_ID')
+        self.NOTION_CLIENT_SECRET = get_env('NOTION_CLIENT_SECRET')
+        self.NOTION_INTEGRATION_TYPE = get_env('NOTION_INTEGRATION_TYPE')
+        self.NOTION_INTERNAL_SECRET = get_env('NOTION_INTERNAL_SECRET')
+
 
 class CloudEditionConfig(Config):
 

api/controllers/console/__init__.py

@@ -5,16 +5,20 @@ from libs.external_api import ExternalApi
 bp = Blueprint('console', __name__, url_prefix='/console/api')
 api = ExternalApi(bp)
 
+# Import other controllers
+from . import setup, version, apikey, admin
+
 # Import app controllers
-from .app import app, site, explore, completion, model_config, statistic, conversation, message
+from .app import app, site, completion, model_config, statistic, conversation, message, generator
 
 # Import auth controllers
-from .auth import login, oauth
+from .auth import login, oauth, data_source_oauth
 
 # Import datasets controllers
-from .datasets import datasets, datasets_document, datasets_segments, file, hit_testing
-
-# Import other controllers
-from . import setup, version, apikey
+from .datasets import datasets, datasets_document, datasets_segments, file, hit_testing, data_source
 
+# Import workspace controllers
 from .workspace import workspace, members, providers, account
+
+# Import explore controllers
+from .explore import installed_app, recommended_app, completion, conversation, message, parameter, saved_message

api/controllers/console/admin.py

@@ -0,0 +1,132 @@
+import os
+from functools import wraps
+
+from flask import request
+from flask_restful import Resource, reqparse
+from werkzeug.exceptions import NotFound, Unauthorized
+
+from controllers.console import api
+from controllers.console.wraps import only_edition_cloud
+from extensions.ext_database import db
+from libs.helper import supported_language
+from models.model import RecommendedApp, App, InstalledApp
+
+
+def admin_required(view):
+    @wraps(view)
+    def decorated(*args, **kwargs):
+        if not os.getenv('ADMIN_API_KEY'):
+            raise Unauthorized('API key is invalid.')
+
+        auth_header = request.headers.get('Authorization')
+        if auth_header is None:
+            raise Unauthorized('Authorization header is missing.')
+
+        if ' ' not in auth_header:
+            raise Unauthorized('Invalid Authorization header format. Expected \'Bearer <api-key>\' format.')
+
+        auth_scheme, auth_token = auth_header.split(None, 1)
+        auth_scheme = auth_scheme.lower()
+
+        if auth_scheme != 'bearer':
+            raise Unauthorized('Invalid Authorization header format. Expected \'Bearer <api-key>\' format.')
+
+        if os.getenv('ADMIN_API_KEY') != auth_token:
+            raise Unauthorized('API key is invalid.')
+
+        return view(*args, **kwargs)
+
+    return decorated
+
+
+class InsertExploreAppListApi(Resource):
+    @only_edition_cloud
+    @admin_required
+    def post(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument('app_id', type=str, required=True, nullable=False, location='json')
+        parser.add_argument('desc', type=str, location='json')
+        parser.add_argument('copyright', type=str, location='json')
+        parser.add_argument('privacy_policy', type=str, location='json')
+        parser.add_argument('language', type=supported_language, required=True, nullable=False, location='json')
+        parser.add_argument('category', type=str, required=True, nullable=False, location='json')
+        parser.add_argument('position', type=int, required=True, nullable=False, location='json')
+        args = parser.parse_args()
+
+        app = App.query.filter(App.id == args['app_id']).first()
+        if not app:
+            raise NotFound('App not found')
+
+        site = app.site
+        if not site:
+            desc = args['desc'] if args['desc'] else ''
+            copy_right = args['copyright'] if args['copyright'] else ''
+            privacy_policy = args['privacy_policy'] if args['privacy_policy'] else ''
+        else:
+            desc = site.description if (site.description if not args['desc'] else args['desc']) else ''
+            copy_right = site.copyright if (site.copyright if not args['copyright'] else args['copyright']) else ''
+            privacy_policy = site.privacy_policy \
+                if (site.privacy_policy if not args['privacy_policy'] else args['privacy_policy']) else ''
+
+        recommended_app = RecommendedApp.query.filter(RecommendedApp.app_id == args['app_id']).first()
+
+        if not recommended_app:
+            recommended_app = RecommendedApp(
+                app_id=app.id,
+                description=desc,
+                copyright=copy_right,
+                privacy_policy=privacy_policy,
+                language=args['language'],
+                category=args['category'],
+                position=args['position']
+            )
+
+            db.session.add(recommended_app)
+
+            app.is_public = True
+            db.session.commit()
+
+            return {'result': 'success'}, 201
+        else:
+            recommended_app.description = desc
+            recommended_app.copyright = copy_right
+            recommended_app.privacy_policy = privacy_policy
+            recommended_app.language = args['language']
+            recommended_app.category = args['category']
+            recommended_app.position = args['position']
+
+            app.is_public = True
+
+            db.session.commit()
+
+            return {'result': 'success'}, 200
+
+
+class InsertExploreAppApi(Resource):
+    @only_edition_cloud
+    @admin_required
+    def delete(self, app_id):
+        recommended_app = RecommendedApp.query.filter(RecommendedApp.app_id == str(app_id)).first()
+        if not recommended_app:
+            return {'result': 'success'}, 204
+
+        app = App.query.filter(App.id == recommended_app.app_id).first()
+        if app:
+            app.is_public = False
+
+        installed_apps = InstalledApp.query.filter(
+            InstalledApp.app_id == recommended_app.app_id,
+            InstalledApp.tenant_id != InstalledApp.app_owner_tenant_id
+        ).all()
+
+        for installed_app in installed_apps:
+            db.session.delete(installed_app)
+
+        db.session.delete(recommended_app)
+        db.session.commit()
+
+        return {'result': 'success'}, 204
+
+
+api.add_resource(InsertExploreAppListApi, '/admin/insert-explore-apps')
+api.add_resource(InsertExploreAppApi, '/admin/insert-explore-apps/<uuid:app_id>')

api/controllers/console/app/__init__.py

@@ -17,6 +17,6 @@ def _get_app(app_id, mode=None):
         raise NotFound("App not found")
 
     if mode and app.mode != mode:
-        raise AppUnavailableError()
+        raise NotFound("The {} app not found".format(mode))
 
     return app

api/controllers/console/app/app.py

@@ -9,18 +9,13 @@ from werkzeug.exceptions import Unauthorized, Forbidden
 
 from constants.model_template import model_templates, demo_model_templates
 from controllers.console import api
-from controllers.console.app.error import AppNotFoundError, ProviderNotInitializeError, ProviderQuotaExceededError, \
-    CompletionRequestError, ProviderModelCurrentlyNotSupportError
+from controllers.console.app.error import AppNotFoundError
 from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
-from core.generator.llm_generator import LLMGenerator
-from core.llm.error import ProviderTokenNotInitError, QuotaExceededError, LLMBadRequestError, LLMAPIConnectionError, \
-    LLMAPIUnavailableError, LLMRateLimitError, LLMAuthorizationError, ModelCurrentlyNotSupportError
 from events.app_event import app_was_created, app_was_deleted
 from libs.helper import TimestampField
 from extensions.ext_database import db
-from models.model import App, AppModelConfig, Site, InstalledApp
-from services.account_service import TenantService
+from models.model import App, AppModelConfig, Site
 from services.app_model_config_service import AppModelConfigService
 
 model_config_fields = {
@@ -220,7 +215,11 @@ class AppTemplateApi(Resource):
         account = current_user
         interface_language = account.interface_language
 
-        return {'data': demo_model_templates.get(interface_language)}
+        templates = demo_model_templates.get(interface_language)
+        if not templates:
+            templates = demo_model_templates.get('en-US')
+
+        return {'data': templates}
 
 
 class AppApi(Resource):
@@ -478,35 +477,6 @@ class AppExport(Resource):
         pass
 
 
-class IntroductionGenerateApi(Resource):
-    @setup_required
-    @login_required
-    @account_initialization_required
-    def post(self):
-        parser = reqparse.RequestParser()
-        parser.add_argument('prompt_template', type=str, required=True, location='json')
-        args = parser.parse_args()
-
-        account = current_user
-
-        try:
-            answer = LLMGenerator.generate_introduction(
-                account.current_tenant_id,
-                args['prompt_template']
-            )
-        except ProviderTokenNotInitError:
-            raise ProviderNotInitializeError()
-        except QuotaExceededError:
-            raise ProviderQuotaExceededError()
-        except ModelCurrentlyNotSupportError:
-            raise ProviderModelCurrentlyNotSupportError()
-        except (LLMBadRequestError, LLMAPIConnectionError, LLMAPIUnavailableError,
-                LLMRateLimitError, LLMAuthorizationError) as e:
-            raise CompletionRequestError(str(e))
-
-        return {'introduction': answer}
-
-
 api.add_resource(AppListApi, '/apps')
 api.add_resource(AppTemplateApi, '/app-templates')
 api.add_resource(AppApi, '/apps/<uuid:app_id>')
@@ -515,4 +485,3 @@ api.add_resource(AppNameApi, '/apps/<uuid:app_id>/name')
 api.add_resource(AppSiteStatus, '/apps/<uuid:app_id>/site-enable')
 api.add_resource(AppApiStatus, '/apps/<uuid:app_id>/api-enable')
 api.add_resource(AppRateLimit, '/apps/<uuid:app_id>/rate-limit')
-api.add_resource(IntroductionGenerateApi, '/introduction-generate')

api/controllers/console/app/conversation.py

@@ -45,7 +45,7 @@ message_detail_fields = {
     'message_tokens': fields.Integer,
     'answer': fields.String,
     'answer_tokens': fields.Integer,
-    'provider_response_latency': fields.Integer,
+    'provider_response_latency': fields.Float,
     'from_source': fields.String,
     'from_end_user_id': fields.String,
     'from_account_id': fields.String,

api/controllers/console/app/error.py

@@ -9,31 +9,33 @@ class AppNotFoundError(BaseHTTPException):
 
 class ProviderNotInitializeError(BaseHTTPException):
     error_code = 'provider_not_initialize'
-    description = "Provider Token not initialize."
+    description = "No valid model provider credentials found. " \
+                  "Please go to Settings -> Model Provider to complete your provider credentials."
     code = 400
 
 
 class ProviderQuotaExceededError(BaseHTTPException):
     error_code = 'provider_quota_exceeded'
-    description = "Provider quota exceeded."
+    description = "Your quota for Dify Hosted OpenAI has been exhausted. " \
+                  "Please go to Settings -> Model Provider to complete your own provider credentials."
     code = 400
 
 
 class ProviderModelCurrentlyNotSupportError(BaseHTTPException):
     error_code = 'model_currently_not_support'
-    description = "GPT-4 currently not support."
+    description = "Dify Hosted OpenAI trial currently not support the GPT-4 model."
     code = 400
 
 
 class ConversationCompletedError(BaseHTTPException):
     error_code = 'conversation_completed'
-    description = "Conversation was completed."
+    description = "The conversation has ended. Please start a new conversation."
     code = 400
 
 
 class AppUnavailableError(BaseHTTPException):
     error_code = 'app_unavailable'
-    description = "App unavailable."
+    description = "App unavailable, please check your app configurations."
     code = 400
 
 
@@ -45,5 +47,5 @@ class CompletionRequestError(BaseHTTPException):
 
 class AppMoreLikeThisDisabledError(BaseHTTPException):
     error_code = 'app_more_like_this_disabled'
-    description = "More like this disabled."
+    description = "The 'More like this' feature is disabled. Please refresh your page."
     code = 403

api/controllers/console/app/explore.py

@@ -1,209 +0,0 @@
-# -*- coding:utf-8 -*-
-from datetime import datetime
-
-from flask_login import login_required, current_user
-from flask_restful import Resource, reqparse, fields, marshal_with, abort, inputs
-from sqlalchemy import and_
-
-from controllers.console import api
-from extensions.ext_database import db
-from models.model import Tenant, App, InstalledApp, RecommendedApp
-from services.account_service import TenantService
-
-app_fields = {
-    'id': fields.String,
-    'name': fields.String,
-    'mode': fields.String,
-    'icon': fields.String,
-    'icon_background': fields.String
-}
-
-installed_app_fields = {
-    'id': fields.String,
-    'app': fields.Nested(app_fields, attribute='app'),
-    'app_owner_tenant_id': fields.String,
-    'is_pinned': fields.Boolean,
-    'last_used_at': fields.DateTime,
-    'editable': fields.Boolean
-}
-
-installed_app_list_fields = {
-    'installed_apps': fields.List(fields.Nested(installed_app_fields))
-}
-
-recommended_app_fields = {
-    'app': fields.Nested(app_fields, attribute='app'),
-    'app_id': fields.String,
-    'description': fields.String(attribute='description'),
-    'copyright': fields.String,
-    'privacy_policy': fields.String,
-    'category': fields.String,
-    'position': fields.Integer,
-    'is_listed': fields.Boolean,
-    'install_count': fields.Integer,
-    'installed': fields.Boolean,
-    'editable': fields.Boolean
-}
-
-recommended_app_list_fields = {
-    'recommended_apps': fields.List(fields.Nested(recommended_app_fields)),
-    'categories': fields.List(fields.String)
-}
-
-
-class InstalledAppsListResource(Resource):
-    @login_required
-    @marshal_with(installed_app_list_fields)
-    def get(self):
-        current_tenant_id = Tenant.query.first().id
-        installed_apps = db.session.query(InstalledApp).filter(
-            InstalledApp.tenant_id == current_tenant_id
-        ).all()
-
-        current_user.role = TenantService.get_user_role(current_user, current_user.current_tenant)
-        installed_apps = [
-            {
-                **installed_app,
-                "editable": current_user.role in ["owner", "admin"],
-            }
-            for installed_app in installed_apps
-        ]
-        installed_apps.sort(key=lambda app: (-app.is_pinned, app.last_used_at))
-
-        return {'installed_apps': installed_apps}
-
-    @login_required
-    def post(self):
-        parser = reqparse.RequestParser()
-        parser.add_argument('app_id', type=str, required=True, help='Invalid app_id')
-        args = parser.parse_args()
-
-        current_tenant_id = Tenant.query.first().id
-        app = App.query.get(args['app_id'])
-        if app is None:
-            abort(404, message='App not found')
-        recommended_app = RecommendedApp.query.filter(RecommendedApp.app_id == args['app_id']).first()
-        if recommended_app is None:
-            abort(404, message='App not found')
-        if not app.is_public:
-            abort(403, message="You can't install a non-public app")
-
-        installed_app = InstalledApp.query.filter(and_(
-            InstalledApp.app_id == args['app_id'],
-            InstalledApp.tenant_id == current_tenant_id
-        )).first()
-
-        if installed_app is None:
-            # todo: position
-            recommended_app.install_count += 1
-
-            new_installed_app = InstalledApp(
-                app_id=args['app_id'],
-                tenant_id=current_tenant_id,
-                is_pinned=False,
-                last_used_at=datetime.utcnow()
-            )
-            db.session.add(new_installed_app)
-            db.session.commit()
-
-        return {'message': 'App installed successfully'}
-
-
-class InstalledAppResource(Resource):
-
-    @login_required
-    def delete(self, installed_app_id):
-
-        installed_app = InstalledApp.query.filter(and_(
-            InstalledApp.id == str(installed_app_id),
-            InstalledApp.tenant_id == current_user.current_tenant_id
-        )).first()
-
-        if installed_app is None:
-            abort(404, message='App not found')
-
-        if installed_app.app_owner_tenant_id == current_user.current_tenant_id:
-            abort(400, message="You can't uninstall an app owned by the current tenant")
-
-        db.session.delete(installed_app)
-        db.session.commit()
-
-        return {'result': 'success', 'message': 'App uninstalled successfully'}
-
-    @login_required
-    def patch(self, installed_app_id):
-        parser = reqparse.RequestParser()
-        parser.add_argument('is_pinned', type=inputs.boolean)
-        args = parser.parse_args()
-
-        current_tenant_id = Tenant.query.first().id
-        installed_app = InstalledApp.query.filter(and_(
-            InstalledApp.id == str(installed_app_id),
-            InstalledApp.tenant_id == current_tenant_id
-        )).first()
-
-        if installed_app is None:
-            abort(404, message='Installed app not found')
-
-        commit_args = False
-        if 'is_pinned' in args:
-            installed_app.is_pinned = args['is_pinned']
-            commit_args = True
-
-        if commit_args:
-            db.session.commit()
-
-        return {'result': 'success', 'message': 'App info updated successfully'}
-
-
-class RecommendedAppsResource(Resource):
-    @login_required
-    @marshal_with(recommended_app_list_fields)
-    def get(self):
-        recommended_apps = db.session.query(RecommendedApp).filter(
-            RecommendedApp.is_listed == True
-        ).all()
-
-        categories = set()
-        current_user.role = TenantService.get_user_role(current_user, current_user.current_tenant)
-        recommended_apps_result = []
-        for recommended_app in recommended_apps:
-            installed = db.session.query(InstalledApp).filter(
-                and_(
-                    InstalledApp.app_id == recommended_app.app_id,
-                    InstalledApp.tenant_id == current_user.current_tenant_id
-                )
-            ).first() is not None
-
-            language_prefix = current_user.interface_language.split('-')[0]
-            desc = None
-            if recommended_app.description:
-                if language_prefix in recommended_app.description:
-                    desc = recommended_app.description[language_prefix]
-                elif 'en' in recommended_app.description:
-                    desc = recommended_app.description['en']
-
-            recommended_app_result = {
-                'id': recommended_app.id,
-                'app': recommended_app.app,
-                'app_id': recommended_app.app_id,
-                'description': desc,
-                'copyright': recommended_app.copyright,
-                'privacy_policy': recommended_app.privacy_policy,
-                'category': recommended_app.category,
-                'position': recommended_app.position,
-                'is_listed': recommended_app.is_listed,
-                'install_count': recommended_app.install_count,
-                'installed': installed,
-                'editable': current_user.role in ['owner', 'admin'],
-            }
-            recommended_apps_result.append(recommended_app_result)
-
-            categories.add(recommended_app.category)  # add category to categories
-
-        return {'recommended_apps': recommended_apps_result, 'categories': list(categories)}
-
-
-api.add_resource(InstalledAppsListResource, '/installed-apps')
-api.add_resource(InstalledAppResource, '/installed-apps/<uuid:installed_app_id>')
-api.add_resource(RecommendedAppsResource, '/explore/apps')

api/controllers/console/app/generator.py

@@ -0,0 +1,75 @@
+from flask_login import login_required, current_user
+from flask_restful import Resource, reqparse
+
+from controllers.console import api
+from controllers.console.app.error import ProviderNotInitializeError, ProviderQuotaExceededError, \
+    CompletionRequestError, ProviderModelCurrentlyNotSupportError
+from controllers.console.setup import setup_required
+from controllers.console.wraps import account_initialization_required
+from core.generator.llm_generator import LLMGenerator
+from core.llm.error import ProviderTokenNotInitError, QuotaExceededError, LLMBadRequestError, LLMAPIConnectionError, \
+    LLMAPIUnavailableError, LLMRateLimitError, LLMAuthorizationError, ModelCurrentlyNotSupportError
+
+
+class IntroductionGenerateApi(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def post(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument('prompt_template', type=str, required=True, location='json')
+        args = parser.parse_args()
+
+        account = current_user
+
+        try:
+            answer = LLMGenerator.generate_introduction(
+                account.current_tenant_id,
+                args['prompt_template']
+            )
+        except ProviderTokenNotInitError:
+            raise ProviderNotInitializeError()
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except (LLMBadRequestError, LLMAPIConnectionError, LLMAPIUnavailableError,
+                LLMRateLimitError, LLMAuthorizationError) as e:
+            raise CompletionRequestError(str(e))
+
+        return {'introduction': answer}
+
+
+class RuleGenerateApi(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def post(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument('audiences', type=str, required=True, nullable=False, location='json')
+        parser.add_argument('hoping_to_solve', type=str, required=True, nullable=False, location='json')
+        args = parser.parse_args()
+
+        account = current_user
+
+        try:
+            rules = LLMGenerator.generate_rule_config(
+                account.current_tenant_id,
+                args['audiences'],
+                args['hoping_to_solve']
+            )
+        except ProviderTokenNotInitError:
+            raise ProviderNotInitializeError()
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except (LLMBadRequestError, LLMAPIConnectionError, LLMAPIUnavailableError,
+                LLMRateLimitError, LLMAuthorizationError) as e:
+            raise CompletionRequestError(str(e))
+
+        return rules
+
+
+api.add_resource(IntroductionGenerateApi, '/introduction-generate')
+api.add_resource(RuleGenerateApi, '/rule-generate')

api/controllers/console/app/message.py

@@ -26,46 +26,46 @@ from services.errors.conversation import ConversationNotExistsError
 from services.errors.message import MessageNotExistsError
 from services.message_service import MessageService
 
+account_fields = {
+    'id': fields.String,
+    'name': fields.String,
+    'email': fields.String
+}
 
-class ChatMessageApi(Resource):
-    account_fields = {
-        'id': fields.String,
-        'name': fields.String,
-        'email': fields.String
-    }
+feedback_fields = {
+    'rating': fields.String,
+    'content': fields.String,
+    'from_source': fields.String,
+    'from_end_user_id': fields.String,
+    'from_account': fields.Nested(account_fields, allow_null=True),
+}
 
-    feedback_fields = {
-        'rating': fields.String,
-        'content': fields.String,
-        'from_source': fields.String,
-        'from_end_user_id': fields.String,
-        'from_account': fields.Nested(account_fields, allow_null=True),
-    }
+annotation_fields = {
+    'content': fields.String,
+    'account': fields.Nested(account_fields, allow_null=True),
+    'created_at': TimestampField
+}
 
-    annotation_fields = {
-        'content': fields.String,
-        'account': fields.Nested(account_fields, allow_null=True),
-        'created_at': TimestampField
-    }
+message_detail_fields = {
+    'id': fields.String,
+    'conversation_id': fields.String,
+    'inputs': fields.Raw,
+    'query': fields.String,
+    'message': fields.Raw,
+    'message_tokens': fields.Integer,
+    'answer': fields.String,
+    'answer_tokens': fields.Integer,
+    'provider_response_latency': fields.Float,
+    'from_source': fields.String,
+    'from_end_user_id': fields.String,
+    'from_account_id': fields.String,
+    'feedbacks': fields.List(fields.Nested(feedback_fields)),
+    'annotation': fields.Nested(annotation_fields, allow_null=True),
+    'created_at': TimestampField
+}
 
-    message_detail_fields = {
-        'id': fields.String,
-        'conversation_id': fields.String,
-        'inputs': fields.Raw,
-        'query': fields.String,
-        'message': fields.Raw,
-        'message_tokens': fields.Integer,
-        'answer': fields.String,
-        'answer_tokens': fields.Integer,
-        'provider_response_latency': fields.Integer,
-        'from_source': fields.String,
-        'from_end_user_id': fields.String,
-        'from_account_id': fields.String,
-        'feedbacks': fields.List(fields.Nested(feedback_fields)),
-        'annotation': fields.Nested(annotation_fields, allow_null=True),
-        'created_at': TimestampField
-    }
 
+class ChatMessageListApi(Resource):
     message_infinite_scroll_pagination_fields = {
         'limit': fields.Integer,
         'has_more': fields.Boolean,
@@ -253,7 +253,8 @@ class MessageMoreLikeThisApi(Resource):
         message_id = str(message_id)
 
         parser = reqparse.RequestParser()
-        parser.add_argument('response_mode', type=str, required=True, choices=['blocking', 'streaming'], location='args')
+        parser.add_argument('response_mode', type=str, required=True, choices=['blocking', 'streaming'],
+                            location='args')
         args = parser.parse_args()
 
         streaming = args['response_mode'] == 'streaming'
@@ -301,7 +302,8 @@ def compact_response(response: Union[dict | Generator]) -> Response:
             except QuotaExceededError:
                 yield "data: " + json.dumps(api.handle_error(ProviderQuotaExceededError()).get_json()) + "\n\n"
             except ModelCurrentlyNotSupportError:
-                yield "data: " + json.dumps(api.handle_error(ProviderModelCurrentlyNotSupportError()).get_json()) + "\n\n"
+                yield "data: " + json.dumps(
+                    api.handle_error(ProviderModelCurrentlyNotSupportError()).get_json()) + "\n\n"
             except (LLMBadRequestError, LLMAPIConnectionError, LLMAPIUnavailableError,
                     LLMRateLimitError, LLMAuthorizationError) as e:
                 yield "data: " + json.dumps(api.handle_error(CompletionRequestError(str(e))).get_json()) + "\n\n"
@@ -353,9 +355,33 @@ class MessageSuggestedQuestionApi(Resource):
         return {'data': questions}
 
 
+class MessageApi(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    @marshal_with(message_detail_fields)
+    def get(self, app_id, message_id):
+        app_id = str(app_id)
+        message_id = str(message_id)
+
+        # get app info
+        app_model = _get_app(app_id, 'chat')
+
+        message = db.session.query(Message).filter(
+            Message.id == message_id,
+            Message.app_id == app_model.id
+        ).first()
+
+        if not message:
+            raise NotFound("Message Not Exists.")
+
+        return message
+
+
 api.add_resource(MessageMoreLikeThisApi, '/apps/<uuid:app_id>/completion-messages/<uuid:message_id>/more-like-this')
 api.add_resource(MessageSuggestedQuestionApi, '/apps/<uuid:app_id>/chat-messages/<uuid:message_id>/suggested-questions')
-api.add_resource(ChatMessageApi, '/apps/<uuid:app_id>/chat-messages', endpoint='chat_messages')
+api.add_resource(ChatMessageListApi, '/apps/<uuid:app_id>/chat-messages', endpoint='console_chat_messages')
 api.add_resource(MessageFeedbackApi, '/apps/<uuid:app_id>/feedbacks')
 api.add_resource(MessageAnnotationApi, '/apps/<uuid:app_id>/annotations')
 api.add_resource(MessageAnnotationCountApi, '/apps/<uuid:app_id>/annotations/count')
+api.add_resource(MessageApi, '/apps/<uuid:app_id>/messages/<uuid:message_id>', endpoint='console_message')

api/controllers/console/app/statistic.py

@@ -1,4 +1,5 @@
 # -*- coding:utf-8 -*-
+from decimal import Decimal
 from datetime import datetime
 
 import pytz
@@ -59,18 +60,20 @@ class DailyConversationStatistic(Resource):
             arg_dict['end'] = end_datetime_utc
 
         sql_query += ' GROUP BY date order by date'
-        rs = db.session.execute(sql_query, arg_dict)
 
-        response_date = []
+        with db.engine.begin() as conn:
+            rs = conn.execute(db.text(sql_query), arg_dict)
+
+        response_data = []
 
         for i in rs:
-            response_date.append({
+            response_data.append({
                 'date': str(i.date),
                 'conversation_count': i.conversation_count
             })
 
         return jsonify({
-            'data': response_date
+            'data': response_data
         })
 
 
@@ -119,18 +122,20 @@ class DailyTerminalsStatistic(Resource):
             arg_dict['end'] = end_datetime_utc
 
         sql_query += ' GROUP BY date order by date'
-        rs = db.session.execute(sql_query, arg_dict)
 
-        response_date = []
+        with db.engine.begin() as conn:
+            rs = conn.execute(db.text(sql_query), arg_dict)
+
+        response_data = []
 
         for i in rs:
-            response_date.append({
+            response_data.append({
                 'date': str(i.date),
                 'terminal_count': i.terminal_count
             })
 
         return jsonify({
-            'data': response_date
+            'data': response_data
         })
 
 
@@ -180,12 +185,14 @@ class DailyTokenCostStatistic(Resource):
             arg_dict['end'] = end_datetime_utc
 
         sql_query += ' GROUP BY date order by date'
-        rs = db.session.execute(sql_query, arg_dict)
 
-        response_date = []
+        with db.engine.begin() as conn:
+            rs = conn.execute(db.text(sql_query), arg_dict)
+
+        response_data = []
 
         for i in rs:
-            response_date.append({
+            response_data.append({
                 'date': str(i.date),
                 'token_count': i.token_count,
                 'total_price': i.total_price,
@@ -193,10 +200,207 @@ class DailyTokenCostStatistic(Resource):
             })
 
         return jsonify({
-            'data': response_date
+            'data': response_data
+        })
+
+
+class AverageSessionInteractionStatistic(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, app_id):
+        account = current_user
+        app_id = str(app_id)
+        app_model = _get_app(app_id, 'chat')
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('start', type=datetime_string('%Y-%m-%d %H:%M'), location='args')
+        parser.add_argument('end', type=datetime_string('%Y-%m-%d %H:%M'), location='args')
+        args = parser.parse_args()
+
+        sql_query = """SELECT date(DATE_TRUNC('day', c.created_at AT TIME ZONE 'UTC' AT TIME ZONE :tz )) AS date, 
+AVG(subquery.message_count) AS interactions
+FROM (SELECT m.conversation_id, COUNT(m.id) AS message_count
+    FROM conversations c
+    JOIN messages m ON c.id = m.conversation_id
+    WHERE c.override_model_configs IS NULL AND c.app_id = :app_id"""
+        arg_dict = {'tz': account.timezone, 'app_id': app_model.id}
+
+        timezone = pytz.timezone(account.timezone)
+        utc_timezone = pytz.utc
+
+        if args['start']:
+            start_datetime = datetime.strptime(args['start'], '%Y-%m-%d %H:%M')
+            start_datetime = start_datetime.replace(second=0)
+
+            start_datetime_timezone = timezone.localize(start_datetime)
+            start_datetime_utc = start_datetime_timezone.astimezone(utc_timezone)
+
+            sql_query += ' and c.created_at >= :start'
+            arg_dict['start'] = start_datetime_utc
+
+        if args['end']:
+            end_datetime = datetime.strptime(args['end'], '%Y-%m-%d %H:%M')
+            end_datetime = end_datetime.replace(second=0)
+
+            end_datetime_timezone = timezone.localize(end_datetime)
+            end_datetime_utc = end_datetime_timezone.astimezone(utc_timezone)
+
+            sql_query += ' and c.created_at < :end'
+            arg_dict['end'] = end_datetime_utc
+
+        sql_query += """
+        GROUP BY m.conversation_id) subquery
+LEFT JOIN conversations c on c.id=subquery.conversation_id
+GROUP BY date
+ORDER BY date"""
+
+        with db.engine.begin() as conn:
+            rs = conn.execute(db.text(sql_query), arg_dict)
+
+        response_data = []
+
+        for i in rs:
+            response_data.append({
+                'date': str(i.date),
+                'interactions': float(i.interactions.quantize(Decimal('0.01')))
+            })
+
+        return jsonify({
+            'data': response_data
+        })
+
+
+class UserSatisfactionRateStatistic(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, app_id):
+        account = current_user
+        app_id = str(app_id)
+        app_model = _get_app(app_id)
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('start', type=datetime_string('%Y-%m-%d %H:%M'), location='args')
+        parser.add_argument('end', type=datetime_string('%Y-%m-%d %H:%M'), location='args')
+        args = parser.parse_args()
+
+        sql_query = '''
+                        SELECT date(DATE_TRUNC('day', m.created_at AT TIME ZONE 'UTC' AT TIME ZONE :tz )) AS date, 
+                            COUNT(m.id) as message_count, COUNT(mf.id) as feedback_count 
+                            FROM messages m
+                            LEFT JOIN message_feedbacks mf on mf.message_id=m.id
+                            WHERE m.app_id = :app_id 
+                        '''
+        arg_dict = {'tz': account.timezone, 'app_id': app_model.id}
+
+        timezone = pytz.timezone(account.timezone)
+        utc_timezone = pytz.utc
+
+        if args['start']:
+            start_datetime = datetime.strptime(args['start'], '%Y-%m-%d %H:%M')
+            start_datetime = start_datetime.replace(second=0)
+
+            start_datetime_timezone = timezone.localize(start_datetime)
+            start_datetime_utc = start_datetime_timezone.astimezone(utc_timezone)
+
+            sql_query += ' and m.created_at >= :start'
+            arg_dict['start'] = start_datetime_utc
+
+        if args['end']:
+            end_datetime = datetime.strptime(args['end'], '%Y-%m-%d %H:%M')
+            end_datetime = end_datetime.replace(second=0)
+
+            end_datetime_timezone = timezone.localize(end_datetime)
+            end_datetime_utc = end_datetime_timezone.astimezone(utc_timezone)
+
+            sql_query += ' and m.created_at < :end'
+            arg_dict['end'] = end_datetime_utc
+
+        sql_query += ' GROUP BY date order by date'
+
+        with db.engine.begin() as conn:
+            rs = conn.execute(db.text(sql_query), arg_dict)
+
+        response_data = []
+
+        for i in rs:
+            response_data.append({
+                'date': str(i.date),
+                'rate': round((i.feedback_count * 1000 / i.message_count) if i.message_count > 0 else 0, 2),
+            })
+
+        return jsonify({
+                'data': response_data
+            })
+
+
+class AverageResponseTimeStatistic(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, app_id):
+        account = current_user
+        app_id = str(app_id)
+        app_model = _get_app(app_id, 'completion')
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('start', type=datetime_string('%Y-%m-%d %H:%M'), location='args')
+        parser.add_argument('end', type=datetime_string('%Y-%m-%d %H:%M'), location='args')
+        args = parser.parse_args()
+
+        sql_query = '''
+                SELECT date(DATE_TRUNC('day', created_at AT TIME ZONE 'UTC' AT TIME ZONE :tz )) AS date, 
+                    AVG(provider_response_latency) as latency
+                    FROM messages
+                    WHERE app_id = :app_id
+                '''
+        arg_dict = {'tz': account.timezone, 'app_id': app_model.id}
+
+        timezone = pytz.timezone(account.timezone)
+        utc_timezone = pytz.utc
+
+        if args['start']:
+            start_datetime = datetime.strptime(args['start'], '%Y-%m-%d %H:%M')
+            start_datetime = start_datetime.replace(second=0)
+
+            start_datetime_timezone = timezone.localize(start_datetime)
+            start_datetime_utc = start_datetime_timezone.astimezone(utc_timezone)
+
+            sql_query += ' and created_at >= :start'
+            arg_dict['start'] = start_datetime_utc
+
+        if args['end']:
+            end_datetime = datetime.strptime(args['end'], '%Y-%m-%d %H:%M')
+            end_datetime = end_datetime.replace(second=0)
+
+            end_datetime_timezone = timezone.localize(end_datetime)
+            end_datetime_utc = end_datetime_timezone.astimezone(utc_timezone)
+
+            sql_query += ' and created_at < :end'
+            arg_dict['end'] = end_datetime_utc
+
+        sql_query += ' GROUP BY date order by date'
+
+        with db.engine.begin() as conn:
+            rs = conn.execute(db.text(sql_query), arg_dict)
+
+        response_data = []
+
+        for i in rs:
+            response_data.append({
+                'date': str(i.date),
+                'latency': round(i.latency * 1000, 4)
+            })
+
+        return jsonify({
+            'data': response_data
         })
 
 
 api.add_resource(DailyConversationStatistic, '/apps/<uuid:app_id>/statistics/daily-conversations')
 api.add_resource(DailyTerminalsStatistic, '/apps/<uuid:app_id>/statistics/daily-end-users')
 api.add_resource(DailyTokenCostStatistic, '/apps/<uuid:app_id>/statistics/token-costs')
+api.add_resource(AverageSessionInteractionStatistic, '/apps/<uuid:app_id>/statistics/average-session-interactions')
+api.add_resource(UserSatisfactionRateStatistic, '/apps/<uuid:app_id>/statistics/user-satisfaction-rate')
+api.add_resource(AverageResponseTimeStatistic, '/apps/<uuid:app_id>/statistics/average-response-time')

api/controllers/console/auth/data_source_oauth.py

@@ -0,0 +1,101 @@
+import logging
+from datetime import datetime
+from typing import Optional
+
+import flask_login
+import requests
+from flask import request, redirect, current_app, session
+from flask_login import current_user, login_required
+from flask_restful import Resource
+from werkzeug.exceptions import Forbidden
+from libs.oauth_data_source import NotionOAuth
+from controllers.console import api
+from ..setup import setup_required
+from ..wraps import account_initialization_required
+
+
+def get_oauth_providers():
+    with current_app.app_context():
+        notion_oauth = NotionOAuth(client_id=current_app.config.get('NOTION_CLIENT_ID'),
+                                   client_secret=current_app.config.get(
+                                       'NOTION_CLIENT_SECRET'),
+                                   redirect_uri=current_app.config.get(
+                                       'CONSOLE_URL') + '/console/api/oauth/data-source/callback/notion')
+
+        OAUTH_PROVIDERS = {
+            'notion': notion_oauth
+        }
+        return OAUTH_PROVIDERS
+
+
+class OAuthDataSource(Resource):
+    def get(self, provider: str):
+        # The role of the current user in the table must be admin or owner
+        if current_user.current_tenant.current_role not in ['admin', 'owner']:
+            raise Forbidden()
+        OAUTH_DATASOURCE_PROVIDERS = get_oauth_providers()
+        with current_app.app_context():
+            oauth_provider = OAUTH_DATASOURCE_PROVIDERS.get(provider)
+            print(vars(oauth_provider))
+        if not oauth_provider:
+            return {'error': 'Invalid provider'}, 400
+        if current_app.config.get('NOTION_INTEGRATION_TYPE') == 'internal':
+            internal_secret = current_app.config.get('NOTION_INTERNAL_SECRET')
+            oauth_provider.save_internal_access_token(internal_secret)
+            return redirect(f'{current_app.config.get("CONSOLE_URL")}?oauth_data_source=success')
+        else:
+            auth_url = oauth_provider.get_authorization_url()
+            return redirect(auth_url)
+
+
+
+
+class OAuthDataSourceCallback(Resource):
+    def get(self, provider: str):
+        OAUTH_DATASOURCE_PROVIDERS = get_oauth_providers()
+        with current_app.app_context():
+            oauth_provider = OAUTH_DATASOURCE_PROVIDERS.get(provider)
+        if not oauth_provider:
+            return {'error': 'Invalid provider'}, 400
+        if 'code' in request.args:
+            code = request.args.get('code')
+            try:
+                oauth_provider.get_access_token(code)
+            except requests.exceptions.HTTPError as e:
+                logging.exception(
+                    f"An error occurred during the OAuthCallback process with {provider}: {e.response.text}")
+                return {'error': 'OAuth data source process failed'}, 400
+
+            return redirect(f'{current_app.config.get("CONSOLE_URL")}?oauth_data_source=success')
+        elif 'error' in request.args:
+            error = request.args.get('error')
+            return redirect(f'{current_app.config.get("CONSOLE_URL")}?oauth_data_source={error}')
+        else:
+            return redirect(f'{current_app.config.get("CONSOLE_URL")}?oauth_data_source=access_denied')
+
+
+class OAuthDataSourceSync(Resource):
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, provider, binding_id):
+        provider = str(provider)
+        binding_id = str(binding_id)
+        OAUTH_DATASOURCE_PROVIDERS = get_oauth_providers()
+        with current_app.app_context():
+            oauth_provider = OAUTH_DATASOURCE_PROVIDERS.get(provider)
+        if not oauth_provider:
+            return {'error': 'Invalid provider'}, 400
+        try:
+            oauth_provider.sync_data_source(binding_id)
+        except requests.exceptions.HTTPError as e:
+            logging.exception(
+                f"An error occurred during the OAuthCallback process with {provider}: {e.response.text}")
+            return {'error': 'OAuth data source process failed'}, 400
+
+        return {'result': 'success'}, 200
+
+
+api.add_resource(OAuthDataSource, '/oauth/data-source/<string:provider>')
+api.add_resource(OAuthDataSourceCallback, '/oauth/data-source/callback/<string:provider>')
+api.add_resource(OAuthDataSourceSync, '/oauth/data-source/<string:provider>/<uuid:binding_id>/sync')

api/controllers/console/datasets/data_source.py

@@ -0,0 +1,303 @@
+import datetime
+import json
+
+from cachetools import TTLCache
+from flask import request, current_app
+from flask_login import login_required, current_user
+from flask_restful import Resource, marshal_with, fields, reqparse, marshal
+from werkzeug.exceptions import NotFound
+
+from controllers.console import api
+from controllers.console.setup import setup_required
+from controllers.console.wraps import account_initialization_required
+from core.data_source.notion import NotionPageReader
+from core.indexing_runner import IndexingRunner
+from extensions.ext_database import db
+from libs.helper import TimestampField
+from libs.oauth_data_source import NotionOAuth
+from models.dataset import Document
+from models.source import DataSourceBinding
+from services.dataset_service import DatasetService, DocumentService
+from tasks.document_indexing_sync_task import document_indexing_sync_task
+
+cache = TTLCache(maxsize=None, ttl=30)
+
+FILE_SIZE_LIMIT = 15 * 1024 * 1024  # 15MB
+ALLOWED_EXTENSIONS = ['txt', 'markdown', 'md', 'pdf', 'html', 'htm']
+PREVIEW_WORDS_LIMIT = 3000
+
+
+class DataSourceApi(Resource):
+    integrate_icon_fields = {
+        'type': fields.String,
+        'url': fields.String,
+        'emoji': fields.String
+    }
+    integrate_page_fields = {
+        'page_name': fields.String,
+        'page_id': fields.String,
+        'page_icon': fields.Nested(integrate_icon_fields, allow_null=True),
+        'parent_id': fields.String,
+        'type': fields.String
+    }
+    integrate_workspace_fields = {
+        'workspace_name': fields.String,
+        'workspace_id': fields.String,
+        'workspace_icon': fields.String,
+        'pages': fields.List(fields.Nested(integrate_page_fields)),
+        'total': fields.Integer
+    }
+    integrate_fields = {
+        'id': fields.String,
+        'provider': fields.String,
+        'created_at': TimestampField,
+        'is_bound': fields.Boolean,
+        'disabled': fields.Boolean,
+        'link': fields.String,
+        'source_info': fields.Nested(integrate_workspace_fields)
+    }
+    integrate_list_fields = {
+        'data': fields.List(fields.Nested(integrate_fields)),
+    }
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    @marshal_with(integrate_list_fields)
+    def get(self):
+        # get workspace data source integrates
+        data_source_integrates = db.session.query(DataSourceBinding).filter(
+            DataSourceBinding.tenant_id == current_user.current_tenant_id,
+            DataSourceBinding.disabled == False
+        ).all()
+
+        base_url = request.url_root.rstrip('/')
+        data_source_oauth_base_path = "/console/api/oauth/data-source"
+        providers = ["notion"]
+
+        integrate_data = []
+        for provider in providers:
+            # existing_integrate = next((ai for ai in data_source_integrates if ai.provider == provider), None)
+            existing_integrates = filter(lambda item: item.provider == provider, data_source_integrates)
+            if existing_integrates:
+                for existing_integrate in list(existing_integrates):
+                    integrate_data.append({
+                        'id': existing_integrate.id,
+                        'provider': provider,
+                        'created_at': existing_integrate.created_at,
+                        'is_bound': True,
+                        'disabled': existing_integrate.disabled,
+                        'source_info': existing_integrate.source_info,
+                        'link': f'{base_url}{data_source_oauth_base_path}/{provider}'
+                })
+            else:
+                integrate_data.append({
+                    'id': None,
+                    'provider': provider,
+                    'created_at': None,
+                    'source_info': None,
+                    'is_bound': False,
+                    'disabled': None,
+                    'link': f'{base_url}{data_source_oauth_base_path}/{provider}'
+                })
+        return {'data': integrate_data}, 200
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def patch(self, binding_id, action):
+        binding_id = str(binding_id)
+        action = str(action)
+        data_source_binding = DataSourceBinding.query.filter_by(
+            id=binding_id
+        ).first()
+        if data_source_binding is None:
+            raise NotFound('Data source binding not found.')
+        # enable binding
+        if action == 'enable':
+            if data_source_binding.disabled:
+                data_source_binding.disabled = False
+                data_source_binding.updated_at = datetime.datetime.utcnow()
+                db.session.add(data_source_binding)
+                db.session.commit()
+            else:
+                raise ValueError('Data source is not disabled.')
+        # disable binding
+        if action == 'disable':
+            if not data_source_binding.disabled:
+                data_source_binding.disabled = True
+                data_source_binding.updated_at = datetime.datetime.utcnow()
+                db.session.add(data_source_binding)
+                db.session.commit()
+            else:
+                raise ValueError('Data source is disabled.')
+        return {'result': 'success'}, 200
+
+
+class DataSourceNotionListApi(Resource):
+    integrate_icon_fields = {
+        'type': fields.String,
+        'url': fields.String,
+        'emoji': fields.String
+    }
+    integrate_page_fields = {
+        'page_name': fields.String,
+        'page_id': fields.String,
+        'page_icon': fields.Nested(integrate_icon_fields, allow_null=True),
+        'is_bound': fields.Boolean,
+        'parent_id': fields.String,
+        'type': fields.String
+    }
+    integrate_workspace_fields = {
+        'workspace_name': fields.String,
+        'workspace_id': fields.String,
+        'workspace_icon': fields.String,
+        'pages': fields.List(fields.Nested(integrate_page_fields))
+    }
+    integrate_notion_info_list_fields = {
+        'notion_info': fields.List(fields.Nested(integrate_workspace_fields)),
+    }
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    @marshal_with(integrate_notion_info_list_fields)
+    def get(self):
+        dataset_id = request.args.get('dataset_id', default=None, type=str)
+        exist_page_ids = []
+        # import notion in the exist dataset
+        if dataset_id:
+            dataset = DatasetService.get_dataset(dataset_id)
+            if not dataset:
+                raise NotFound('Dataset not found.')
+            if dataset.data_source_type != 'notion_import':
+                raise ValueError('Dataset is not notion type.')
+            documents = Document.query.filter_by(
+                dataset_id=dataset_id,
+                tenant_id=current_user.current_tenant_id,
+                data_source_type='notion_import',
+                enabled=True
+            ).all()
+            if documents:
+                for document in documents:
+                    data_source_info = json.loads(document.data_source_info)
+                    exist_page_ids.append(data_source_info['notion_page_id'])
+        # get all authorized pages
+        data_source_bindings = DataSourceBinding.query.filter_by(
+            tenant_id=current_user.current_tenant_id,
+            provider='notion',
+            disabled=False
+        ).all()
+        if not data_source_bindings:
+            return {
+                'notion_info': []
+            }, 200
+        pre_import_info_list = []
+        for data_source_binding in data_source_bindings:
+            source_info = data_source_binding.source_info
+            pages = source_info['pages']
+            # Filter out already bound pages
+            for page in pages:
+                if page['page_id'] in exist_page_ids:
+                    page['is_bound'] = True
+                else:
+                    page['is_bound'] = False
+            pre_import_info = {
+                'workspace_name': source_info['workspace_name'],
+                'workspace_icon': source_info['workspace_icon'],
+                'workspace_id': source_info['workspace_id'],
+                'pages': pages,
+            }
+            pre_import_info_list.append(pre_import_info)
+        return {
+            'notion_info': pre_import_info_list
+        }, 200
+
+
+class DataSourceNotionApi(Resource):
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, workspace_id, page_id, page_type):
+        workspace_id = str(workspace_id)
+        page_id = str(page_id)
+        data_source_binding = DataSourceBinding.query.filter(
+            db.and_(
+                DataSourceBinding.tenant_id == current_user.current_tenant_id,
+                DataSourceBinding.provider == 'notion',
+                DataSourceBinding.disabled == False,
+                DataSourceBinding.source_info['workspace_id'] == f'"{workspace_id}"'
+            )
+        ).first()
+        if not data_source_binding:
+            raise NotFound('Data source binding not found.')
+        reader = NotionPageReader(integration_token=data_source_binding.access_token)
+        if page_type == 'page':
+            page_content = reader.read_page(page_id)
+        elif page_type == 'database':
+            page_content = reader.query_database_data(page_id)
+        else:
+            page_content = ""
+        return {
+            'content': page_content
+        }, 200
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def post(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument('notion_info_list', type=list, required=True, nullable=True, location='json')
+        parser.add_argument('process_rule', type=dict, required=True, nullable=True, location='json')
+        args = parser.parse_args()
+        # validate args
+        DocumentService.estimate_args_validate(args)
+        indexing_runner = IndexingRunner()
+        response = indexing_runner.notion_indexing_estimate(args['notion_info_list'], args['process_rule'])
+        return response, 200
+
+
+class DataSourceNotionDatasetSyncApi(Resource):
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, dataset_id):
+        dataset_id_str = str(dataset_id)
+        dataset = DatasetService.get_dataset(dataset_id_str)
+        if dataset is None:
+            raise NotFound("Dataset not found.")
+
+        documents = DocumentService.get_document_by_dataset_id(dataset_id_str)
+        for document in documents:
+            document_indexing_sync_task.delay(dataset_id_str, document.id)
+        return 200
+
+
+class DataSourceNotionDocumentSyncApi(Resource):
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, dataset_id, document_id):
+        dataset_id_str = str(dataset_id)
+        document_id_str = str(document_id)
+        dataset = DatasetService.get_dataset(dataset_id_str)
+        if dataset is None:
+            raise NotFound("Dataset not found.")
+
+        document = DocumentService.get_document(dataset_id_str, document_id_str)
+        if document is None:
+            raise NotFound("Document not found.")
+        document_indexing_sync_task.delay(dataset_id_str, document_id_str)
+        return 200
+
+
+api.add_resource(DataSourceApi, '/data-source/integrates', '/data-source/integrates/<uuid:binding_id>/<string:action>')
+api.add_resource(DataSourceNotionListApi, '/notion/pre-import/pages')
+api.add_resource(DataSourceNotionApi,
+                 '/notion/workspaces/<uuid:workspace_id>/pages/<uuid:page_id>/<string:page_type>/preview',
+                 '/datasets/notion-indexing-estimate')
+api.add_resource(DataSourceNotionDatasetSyncApi, '/datasets/<uuid:dataset_id>/notion/sync')
+api.add_resource(DataSourceNotionDocumentSyncApi, '/datasets/<uuid:dataset_id>/documents/<uuid:document_id>/notion/sync')

api/controllers/console/datasets/datasets.py

@@ -12,8 +12,9 @@ from controllers.console.wraps import account_initialization_required
 from core.indexing_runner import IndexingRunner
 from libs.helper import TimestampField
 from extensions.ext_database import db
+from models.dataset import DocumentSegment, Document
 from models.model import UploadFile
-from services.dataset_service import DatasetService
+from services.dataset_service import DatasetService, DocumentService
 
 dataset_detail_fields = {
     'id': fields.String,
@@ -50,8 +51,8 @@ def _validate_name(name):
 
 
 def _validate_description_length(description):
-    if len(description) > 200:
-        raise ValueError('Description cannot exceed 200 characters.')
+    if len(description) > 400:
+        raise ValueError('Description cannot exceed 400 characters.')
     return description
 
 
@@ -217,17 +218,31 @@ class DatasetIndexingEstimateApi(Resource):
     @login_required
     @account_initialization_required
     def post(self):
-        segment_rule = request.get_json()
-        file_detail = db.session.query(UploadFile).filter(
-            UploadFile.tenant_id == current_user.current_tenant_id,
-            UploadFile.id == segment_rule["file_id"]
-        ).first()
+        parser = reqparse.RequestParser()
+        parser.add_argument('info_list', type=dict, required=True, nullable=True, location='json')
+        parser.add_argument('process_rule', type=dict, required=True, nullable=True, location='json')
+        args = parser.parse_args()
+        # validate args
+        DocumentService.estimate_args_validate(args)
+        if args['info_list']['data_source_type'] == 'upload_file':
+            file_ids = args['info_list']['file_info_list']['file_ids']
+            file_details = db.session.query(UploadFile).filter(
+                UploadFile.tenant_id == current_user.current_tenant_id,
+                UploadFile.id.in_(file_ids)
+            ).all()
 
-        if file_detail is None:
-            raise NotFound("File not found.")
+            if file_details is None:
+                raise NotFound("File not found.")
 
-        indexing_runner = IndexingRunner()
-        response = indexing_runner.indexing_estimate(file_detail, segment_rule['process_rule'])
+            indexing_runner = IndexingRunner()
+            response = indexing_runner.file_indexing_estimate(file_details, args['process_rule'])
+        elif args['info_list']['data_source_type'] == 'notion_import':
+
+            indexing_runner = IndexingRunner()
+            response = indexing_runner.notion_indexing_estimate(args['info_list']['notion_info_list'],
+                                                                args['process_rule'])
+        else:
+            raise ValueError('Data source type not support')
         return response, 200
 
 
@@ -274,8 +289,54 @@ class DatasetRelatedAppListApi(Resource):
         }, 200
 
 
+class DatasetIndexingStatusApi(Resource):
+    document_status_fields = {
+        'id': fields.String,
+        'indexing_status': fields.String,
+        'processing_started_at': TimestampField,
+        'parsing_completed_at': TimestampField,
+        'cleaning_completed_at': TimestampField,
+        'splitting_completed_at': TimestampField,
+        'completed_at': TimestampField,
+        'paused_at': TimestampField,
+        'error': fields.String,
+        'stopped_at': TimestampField,
+        'completed_segments': fields.Integer,
+        'total_segments': fields.Integer,
+    }
+
+    document_status_fields_list = {
+        'data': fields.List(fields.Nested(document_status_fields))
+    }
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, dataset_id):
+        dataset_id = str(dataset_id)
+        documents = db.session.query(Document).filter(
+            Document.dataset_id == dataset_id,
+            Document.tenant_id == current_user.current_tenant_id
+        ).all()
+        documents_status = []
+        for document in documents:
+            completed_segments = DocumentSegment.query.filter(DocumentSegment.completed_at.isnot(None),
+                                                              DocumentSegment.document_id == str(document.id),
+                                                              DocumentSegment.status != 're_segment').count()
+            total_segments = DocumentSegment.query.filter(DocumentSegment.document_id == str(document.id),
+                                                          DocumentSegment.status != 're_segment').count()
+            document.completed_segments = completed_segments
+            document.total_segments = total_segments
+            documents_status.append(marshal(document, self.document_status_fields))
+        data = {
+            'data': documents_status
+        }
+        return data
+
+
 api.add_resource(DatasetListApi, '/datasets')
 api.add_resource(DatasetApi, '/datasets/<uuid:dataset_id>')
 api.add_resource(DatasetQueryApi, '/datasets/<uuid:dataset_id>/queries')
-api.add_resource(DatasetIndexingEstimateApi, '/datasets/file-indexing-estimate')
+api.add_resource(DatasetIndexingEstimateApi, '/datasets/indexing-estimate')
 api.add_resource(DatasetRelatedAppListApi, '/datasets/<uuid:dataset_id>/related-apps')
+api.add_resource(DatasetIndexingStatusApi, '/datasets/<uuid:dataset_id>/indexing-status')

api/controllers/console/datasets/datasets_document.py

@@ -1,6 +1,7 @@
 # -*- coding:utf-8 -*-
 import random
 from datetime import datetime
+from typing import List
 
 from flask import request
 from flask_login import login_required, current_user
@@ -10,13 +11,14 @@ from werkzeug.exceptions import NotFound, Forbidden
 
 import services
 from controllers.console import api
-from controllers.console.app.error import ProviderNotInitializeError
+from controllers.console.app.error import ProviderNotInitializeError, ProviderQuotaExceededError, \
+    ProviderModelCurrentlyNotSupportError
 from controllers.console.datasets.error import DocumentAlreadyFinishedError, InvalidActionError, DocumentIndexingError, \
     InvalidMetadataError, ArchivedDocumentImmutableError
 from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
 from core.indexing_runner import IndexingRunner
-from core.llm.error import ProviderTokenNotInitError
+from core.llm.error import ProviderTokenNotInitError, QuotaExceededError, ModelCurrentlyNotSupportError
 from extensions.ext_redis import redis_client
 from libs.helper import TimestampField
 from extensions.ext_database import db
@@ -60,6 +62,29 @@ document_fields = {
     'hit_count': fields.Integer,
 }
 
+document_with_segments_fields = {
+    'id': fields.String,
+    'position': fields.Integer,
+    'data_source_type': fields.String,
+    'data_source_info': fields.Raw(attribute='data_source_info_dict'),
+    'dataset_process_rule_id': fields.String,
+    'name': fields.String,
+    'created_from': fields.String,
+    'created_by': fields.String,
+    'created_at': TimestampField,
+    'tokens': fields.Integer,
+    'indexing_status': fields.String,
+    'error': fields.String,
+    'enabled': fields.Boolean,
+    'disabled_at': TimestampField,
+    'disabled_by': fields.String,
+    'archived': fields.Boolean,
+    'display_status': fields.String,
+    'word_count': fields.Integer,
+    'hit_count': fields.Integer,
+    'completed_segments': fields.Integer,
+    'total_segments': fields.Integer
+}
 
 class DocumentResource(Resource):
     def get_document(self, dataset_id: str, document_id: str) -> Document:
@@ -82,6 +107,23 @@ class DocumentResource(Resource):
 
         return document
 
+    def get_batch_documents(self, dataset_id: str, batch: str) -> List[Document]:
+        dataset = DatasetService.get_dataset(dataset_id)
+        if not dataset:
+            raise NotFound('Dataset not found.')
+
+        try:
+            DatasetService.check_dataset_permission(dataset, current_user)
+        except services.errors.account.NoPermissionError as e:
+            raise Forbidden(str(e))
+
+        documents = DocumentService.get_batch_documents(dataset_id, batch)
+
+        if not documents:
+            raise NotFound('Documents not found.')
+
+        return documents
+
 
 class GetProcessRuleApi(Resource):
     @setup_required
@@ -131,9 +173,9 @@ class DatasetDocumentListApi(Resource):
         dataset_id = str(dataset_id)
         page = request.args.get('page', default=1, type=int)
         limit = request.args.get('limit', default=20, type=int)
-        search = request.args.get('search', default=None, type=str)
+        search = request.args.get('keyword', default=None, type=str)
         sort = request.args.get('sort', default='-created_at', type=str)
-
+        fetch = request.args.get('fetch', default=False, type=bool)
         dataset = DatasetService.get_dataset(dataset_id)
         if not dataset:
             raise NotFound('Dataset not found.')
@@ -172,9 +214,20 @@ class DatasetDocumentListApi(Resource):
         paginated_documents = query.paginate(
             page=page, per_page=limit, max_per_page=100, error_out=False)
         documents = paginated_documents.items
-
+        if fetch:
+            for document in documents:
+                completed_segments = DocumentSegment.query.filter(DocumentSegment.completed_at.isnot(None),
+                                                                  DocumentSegment.document_id == str(document.id),
+                                                                  DocumentSegment.status != 're_segment').count()
+                total_segments = DocumentSegment.query.filter(DocumentSegment.document_id == str(document.id),
+                                                              DocumentSegment.status != 're_segment').count()
+                document.completed_segments = completed_segments
+                document.total_segments = total_segments
+            data = marshal(documents, document_with_segments_fields)
+        else:
+            data = marshal(documents, document_fields)
         response = {
-            'data': marshal(documents, document_fields),
+            'data': data,
             'has_more': len(documents) == limit,
             'limit': limit,
             'total': paginated_documents.total,
@@ -183,10 +236,15 @@ class DatasetDocumentListApi(Resource):
 
         return response
 
+    documents_and_batch_fields = {
+        'documents': fields.List(fields.Nested(document_fields)),
+        'batch': fields.String
+    }
+
     @setup_required
     @login_required
     @account_initialization_required
-    @marshal_with(document_fields)
+    @marshal_with(documents_and_batch_fields)
     def post(self, dataset_id):
         dataset_id = str(dataset_id)
 
@@ -207,9 +265,10 @@ class DatasetDocumentListApi(Resource):
         parser = reqparse.RequestParser()
         parser.add_argument('indexing_technique', type=str, choices=Dataset.INDEXING_TECHNIQUE_LIST, nullable=False,
                             location='json')
-        parser.add_argument('data_source', type=dict, required=True, nullable=True, location='json')
-        parser.add_argument('process_rule', type=dict, required=True, nullable=True, location='json')
+        parser.add_argument('data_source', type=dict, required=False, location='json')
+        parser.add_argument('process_rule', type=dict, required=False, location='json')
         parser.add_argument('duplicate', type=bool, nullable=False, location='json')
+        parser.add_argument('original_document_id', type=str, required=False, location='json')
         args = parser.parse_args()
 
         if not dataset.indexing_technique and not args['indexing_technique']:
@@ -219,17 +278,25 @@ class DatasetDocumentListApi(Resource):
         DocumentService.document_create_args_validate(args)
 
         try:
-            document = DocumentService.save_document_with_dataset_id(dataset, args, current_user)
+            documents, batch = DocumentService.save_document_with_dataset_id(dataset, args, current_user)
         except ProviderTokenNotInitError:
             raise ProviderNotInitializeError()
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
 
-        return document
+        return {
+            'documents': documents,
+            'batch': batch
+        }
 
 
 class DatasetInitApi(Resource):
     dataset_and_document_fields = {
         'dataset': fields.Nested(dataset_fields),
-        'document': fields.Nested(document_fields)
+        'documents': fields.List(fields.Nested(document_fields)),
+        'batch': fields.String
     }
 
     @setup_required
@@ -252,17 +319,22 @@ class DatasetInitApi(Resource):
         DocumentService.document_create_args_validate(args)
 
         try:
-            dataset, document = DocumentService.save_document_without_dataset_id(
+            dataset, documents, batch = DocumentService.save_document_without_dataset_id(
                 tenant_id=current_user.current_tenant_id,
                 document_data=args,
                 account=current_user
             )
         except ProviderTokenNotInitError:
             raise ProviderNotInitializeError()
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
 
         response = {
             'dataset': dataset,
-            'document': document
+            'documents': documents,
+            'batch': batch
         }
 
         return response
@@ -307,11 +379,122 @@ class DocumentIndexingEstimateApi(DocumentResource):
                     raise NotFound('File not found.')
 
                 indexing_runner = IndexingRunner()
-                response = indexing_runner.indexing_estimate(file, data_process_rule_dict)
+
+                response = indexing_runner.file_indexing_estimate([file], data_process_rule_dict)
 
         return response
 
 
+class DocumentBatchIndexingEstimateApi(DocumentResource):
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, dataset_id, batch):
+        dataset_id = str(dataset_id)
+        batch = str(batch)
+        dataset = DatasetService.get_dataset(dataset_id)
+        if dataset is None:
+            raise NotFound("Dataset not found.")
+        documents = self.get_batch_documents(dataset_id, batch)
+        response = {
+            "tokens": 0,
+            "total_price": 0,
+            "currency": "USD",
+            "total_segments": 0,
+            "preview": []
+        }
+        if not documents:
+            return response
+        data_process_rule = documents[0].dataset_process_rule
+        data_process_rule_dict = data_process_rule.to_dict()
+        info_list = []
+        for document in documents:
+            if document.indexing_status in ['completed', 'error']:
+                raise DocumentAlreadyFinishedError()
+            data_source_info = document.data_source_info_dict
+            # format document files info
+            if data_source_info and 'upload_file_id' in data_source_info:
+                file_id = data_source_info['upload_file_id']
+                info_list.append(file_id)
+            # format document notion info
+            elif data_source_info and 'notion_workspace_id' in data_source_info and 'notion_page_id' in data_source_info:
+                pages = []
+                page = {
+                    'page_id': data_source_info['notion_page_id'],
+                    'type': data_source_info['type']
+                }
+                pages.append(page)
+                notion_info = {
+                    'workspace_id': data_source_info['notion_workspace_id'],
+                    'pages': pages
+                }
+                info_list.append(notion_info)
+
+        if dataset.data_source_type == 'upload_file':
+            file_details = db.session.query(UploadFile).filter(
+                UploadFile.tenant_id == current_user.current_tenant_id,
+                UploadFile.id in info_list
+            ).all()
+
+            if file_details is None:
+                raise NotFound("File not found.")
+
+            indexing_runner = IndexingRunner()
+            response = indexing_runner.file_indexing_estimate(file_details, data_process_rule_dict)
+        elif dataset.data_source_type:
+
+            indexing_runner = IndexingRunner()
+            response = indexing_runner.notion_indexing_estimate(info_list,
+                                                                data_process_rule_dict)
+        else:
+            raise ValueError('Data source type not support')
+        return response
+
+
+class DocumentBatchIndexingStatusApi(DocumentResource):
+    document_status_fields = {
+        'id': fields.String,
+        'indexing_status': fields.String,
+        'processing_started_at': TimestampField,
+        'parsing_completed_at': TimestampField,
+        'cleaning_completed_at': TimestampField,
+        'splitting_completed_at': TimestampField,
+        'completed_at': TimestampField,
+        'paused_at': TimestampField,
+        'error': fields.String,
+        'stopped_at': TimestampField,
+        'completed_segments': fields.Integer,
+        'total_segments': fields.Integer,
+    }
+
+    document_status_fields_list = {
+        'data': fields.List(fields.Nested(document_status_fields))
+    }
+
+    @setup_required
+    @login_required
+    @account_initialization_required
+    def get(self, dataset_id, batch):
+        dataset_id = str(dataset_id)
+        batch = str(batch)
+        documents = self.get_batch_documents(dataset_id, batch)
+        documents_status = []
+        for document in documents:
+            completed_segments = DocumentSegment.query.filter(DocumentSegment.completed_at.isnot(None),
+                                                              DocumentSegment.document_id == str(document.id),
+                                                              DocumentSegment.status != 're_segment').count()
+            total_segments = DocumentSegment.query.filter(DocumentSegment.document_id == str(document.id),
+                                                          DocumentSegment.status != 're_segment').count()
+            document.completed_segments = completed_segments
+            document.total_segments = total_segments
+            documents_status.append(marshal(document, self.document_status_fields))
+        data = {
+            'data': documents_status
+        }
+        return data
+
+
 class DocumentIndexingStatusApi(DocumentResource):
     document_status_fields = {
         'id': fields.String,
@@ -338,10 +521,12 @@ class DocumentIndexingStatusApi(DocumentResource):
 
         completed_segments = DocumentSegment.query \
             .filter(DocumentSegment.completed_at.isnot(None),
-                    DocumentSegment.document_id == str(document_id)) \
+                    DocumentSegment.document_id == str(document_id),
+                    DocumentSegment.status != 're_segment') \
             .count()
         total_segments = DocumentSegment.query \
-            .filter_by(document_id=str(document_id)) \
+            .filter(DocumentSegment.document_id == str(document_id),
+                    DocumentSegment.status != 're_segment') \
             .count()
 
         document.completed_segments = completed_segments
@@ -396,7 +581,7 @@ class DocumentDetailApi(DocumentResource):
                 'disabled_by': document.disabled_by,
                 'archived': document.archived,
                 'segment_count': document.segment_count,
-                'average_segment_length':   document.average_segment_length,
+                'average_segment_length': document.average_segment_length,
                 'hit_count': document.hit_count,
                 'display_status': document.display_status
             }
@@ -416,7 +601,7 @@ class DocumentDetailApi(DocumentResource):
                 'created_at': document.created_at.timestamp(),
                 'tokens': document.tokens,
                 'indexing_status': document.indexing_status,
-                'completed_at': int(document.completed_at.timestamp())if document.completed_at else None,
+                'completed_at': int(document.completed_at.timestamp()) if document.completed_at else None,
                 'updated_at': int(document.updated_at.timestamp()) if document.updated_at else None,
                 'indexing_latency': document.indexing_latency,
                 'error': document.error,
@@ -567,6 +752,8 @@ class DocumentStatusApi(DocumentResource):
             return {'result': 'success'}, 200
 
         elif action == "disable":
+            if not document.completed_at or document.indexing_status != 'completed':
+                raise InvalidActionError('Document is not completed.')
             if not document.enabled:
                 raise InvalidActionError('Document already disabled.')
 
@@ -666,6 +853,10 @@ api.add_resource(DatasetInitApi,
                  '/datasets/init')
 api.add_resource(DocumentIndexingEstimateApi,
                  '/datasets/<uuid:dataset_id>/documents/<uuid:document_id>/indexing-estimate')
+api.add_resource(DocumentBatchIndexingEstimateApi,
+                 '/datasets/<uuid:dataset_id>/batch/<string:batch>/indexing-estimate')
+api.add_resource(DocumentBatchIndexingStatusApi,
+                 '/datasets/<uuid:dataset_id>/batch/<string:batch>/indexing-status')
 api.add_resource(DocumentIndexingStatusApi,
                  '/datasets/<uuid:dataset_id>/documents/<uuid:document_id>/indexing-status')
 api.add_resource(DocumentDetailApi,

api/controllers/console/datasets/datasets_segments.py

@@ -78,12 +78,14 @@ class DatasetDocumentSegmentListApi(Resource):
         parser.add_argument('hit_count_gte', type=int,
                             default=None, location='args')
         parser.add_argument('enabled', type=str, default='all', location='args')
+        parser.add_argument('keyword', type=str, default=None, location='args')
         args = parser.parse_args()
 
         last_id = args['last_id']
         limit = min(args['limit'], 100)
         status_list = args['status']
         hit_count_gte = args['hit_count_gte']
+        keyword = args['keyword']
 
         query = DocumentSegment.query.filter(
             DocumentSegment.document_id == str(document_id),
@@ -104,6 +106,9 @@ class DatasetDocumentSegmentListApi(Resource):
         if hit_count_gte is not None:
             query = query.filter(DocumentSegment.hit_count >= hit_count_gte)
 
+        if keyword:
+            query = query.where(DocumentSegment.content.ilike(f'%{keyword}%'))
+
         if args['enabled'].lower() != 'all':
             if args['enabled'].lower() == 'true':
                 query = query.filter(DocumentSegment.enabled == True)

api/controllers/console/datasets/error.py

@@ -3,7 +3,7 @@ from libs.exception import BaseHTTPException
 
 class NoFileUploadedError(BaseHTTPException):
     error_code = 'no_file_uploaded'
-    description = "No file uploaded."
+    description = "Please upload your file."
     code = 400
 
 
@@ -27,25 +27,25 @@ class UnsupportedFileTypeError(BaseHTTPException):
 
 class HighQualityDatasetOnlyError(BaseHTTPException):
     error_code = 'high_quality_dataset_only'
-    description = "High quality dataset only."
+    description = "Current operation only supports 'high-quality' datasets."
     code = 400
 
 
 class DatasetNotInitializedError(BaseHTTPException):
     error_code = 'dataset_not_initialized'
-    description = "Dataset not initialized."
+    description = "The dataset is still being initialized or indexing. Please wait a moment."
     code = 400
 
 
 class ArchivedDocumentImmutableError(BaseHTTPException):
     error_code = 'archived_document_immutable'
-    description = "Cannot process an archived document."
+    description = "The archived document is not editable."
     code = 403
 
 
 class DatasetNameDuplicateError(BaseHTTPException):
     error_code = 'dataset_name_duplicate'
-    description = "Dataset name already exists."
+    description = "The dataset name already exists. Please modify your dataset name."
     code = 409
 
 
@@ -57,17 +57,17 @@ class InvalidActionError(BaseHTTPException):
 
 class DocumentAlreadyFinishedError(BaseHTTPException):
     error_code = 'document_already_finished'
-    description = "Document already finished."
+    description = "The document has been processed. Please refresh the page or go to the document details."
     code = 400
 
 
 class DocumentIndexingError(BaseHTTPException):
     error_code = 'document_indexing'
-    description = "Document indexing."
+    description = "The document is being processed and cannot be edited."
     code = 400
 
 
 class InvalidMetadataError(BaseHTTPException):
     error_code = 'invalid_metadata'
-    description = "Invalid metadata."
+    description = "The metadata content is incorrect. Please check and verify."
     code = 400

api/controllers/console/datasets/file.py

@@ -1,6 +1,7 @@
 import datetime
 import hashlib
 import tempfile
+import chardet
 import time
 import uuid
 from pathlib import Path
@@ -18,6 +19,7 @@ from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
 from core.index.readers.html_parser import HTMLParser
 from core.index.readers.pdf_parser import PDFParser
+from core.index.readers.xlsx_parser import XLSXParser
 from extensions.ext_storage import storage
 from libs.helper import TimestampField
 from extensions.ext_database import db
@@ -26,7 +28,7 @@ from models.model import UploadFile
 cache = TTLCache(maxsize=None, ttl=30)
 
 FILE_SIZE_LIMIT = 15 * 1024 * 1024  # 15MB
-ALLOWED_EXTENSIONS = ['txt', 'markdown', 'md', 'pdf', 'html', 'htm']
+ALLOWED_EXTENSIONS = ['txt', 'markdown', 'md', 'pdf', 'html', 'htm', 'xlsx']
 PREVIEW_WORDS_LIMIT = 3000
 
 
@@ -133,11 +135,18 @@ class FilePreviewApi(Resource):
                 # Use BeautifulSoup to extract text
                 parser = HTMLParser()
                 text = parser.parse_file(Path(filepath))
+            elif extension == 'xlsx':
+                parser = XLSXParser()
+                text = parser.parse_file(filepath)
             else:
                 # ['txt', 'markdown', 'md']
                 with open(filepath, "rb") as fp:
                     data = fp.read()
-                    text = data.decode(encoding='utf-8').strip() if data else ''
+                    encoding = chardet.detect(data)['encoding']
+                    if encoding:
+                        text = data.decode(encoding=encoding).strip() if data else ''
+                    else:
+                        text = data.decode(encoding='utf-8').strip() if data else ''
 
         text = text[0:PREVIEW_WORDS_LIMIT] if text else ''
         return {'content': text}

api/controllers/console/datasets/hit_testing.py

@@ -6,9 +6,12 @@ from werkzeug.exceptions import InternalServerError, NotFound, Forbidden
 
 import services
 from controllers.console import api
+from controllers.console.app.error import ProviderNotInitializeError, ProviderQuotaExceededError, \
+    ProviderModelCurrentlyNotSupportError
 from controllers.console.datasets.error import HighQualityDatasetOnlyError, DatasetNotInitializedError
 from controllers.console.setup import setup_required
 from controllers.console.wraps import account_initialization_required
+from core.llm.error import ProviderTokenNotInitError, QuotaExceededError, ModelCurrentlyNotSupportError
 from libs.helper import TimestampField
 from services.dataset_service import DatasetService
 from services.hit_testing_service import HitTestingService
@@ -92,6 +95,12 @@ class HitTestingApi(Resource):
             return {"query": response['query'], 'records': marshal(response['records'], hit_testing_record_fields)}
         except services.errors.index.IndexNotInitializedError:
             raise DatasetNotInitializedError()
+        except ProviderTokenNotInitError:
+            raise ProviderNotInitializeError()
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
         except Exception as e:
             logging.exception("Hit testing failed.")
             raise InternalServerError(str(e))

api/controllers/console/error.py

@@ -3,13 +3,14 @@ from libs.exception import BaseHTTPException
 
 class AlreadySetupError(BaseHTTPException):
     error_code = 'already_setup'
-    description = "Application already setup."
+    description = "Dify has been successfully installed. Please refresh the page or return to the dashboard homepage."
     code = 403
 
 
 class NotSetupError(BaseHTTPException):
     error_code = 'not_setup'
-    description = "Application not setup."
+    description = "Dify has not been initialized and installed yet. " \
+                  "Please proceed with the initialization and installation process first."
     code = 401
 
 

api/controllers/console/explore/completion.py

@@ -0,0 +1,180 @@
+# -*- coding:utf-8 -*-
+import json
+import logging
+from typing import Generator, Union
+
+from flask import Response, stream_with_context
+from flask_login import current_user
+from flask_restful import reqparse
+from werkzeug.exceptions import InternalServerError, NotFound
+
+import services
+from controllers.console import api
+from controllers.console.app.error import ConversationCompletedError, AppUnavailableError, ProviderNotInitializeError, \
+    ProviderQuotaExceededError, ProviderModelCurrentlyNotSupportError, CompletionRequestError
+from controllers.console.explore.error import NotCompletionAppError, NotChatAppError
+from controllers.console.explore.wraps import InstalledAppResource
+from core.conversation_message_task import PubHandler
+from core.llm.error import LLMBadRequestError, LLMAPIUnavailableError, LLMAuthorizationError, LLMAPIConnectionError, \
+    LLMRateLimitError, ProviderTokenNotInitError, QuotaExceededError, ModelCurrentlyNotSupportError
+from libs.helper import uuid_value
+from services.completion_service import CompletionService
+
+
+# define completion api for user
+class CompletionApi(InstalledAppResource):
+
+    def post(self, installed_app):
+        app_model = installed_app.app
+        if app_model.mode != 'completion':
+            raise NotCompletionAppError()
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('inputs', type=dict, required=True, location='json')
+        parser.add_argument('query', type=str, location='json')
+        parser.add_argument('response_mode', type=str, choices=['blocking', 'streaming'], location='json')
+        args = parser.parse_args()
+
+        streaming = args['response_mode'] == 'streaming'
+
+        try:
+            response = CompletionService.completion(
+                app_model=app_model,
+                user=current_user,
+                args=args,
+                from_source='console',
+                streaming=streaming
+            )
+
+            return compact_response(response)
+        except services.errors.conversation.ConversationNotExistsError:
+            raise NotFound("Conversation Not Exists.")
+        except services.errors.conversation.ConversationCompletedError:
+            raise ConversationCompletedError()
+        except services.errors.app_model_config.AppModelConfigBrokenError:
+            logging.exception("App model config broken.")
+            raise AppUnavailableError()
+        except ProviderTokenNotInitError:
+            raise ProviderNotInitializeError()
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except (LLMBadRequestError, LLMAPIConnectionError, LLMAPIUnavailableError,
+                LLMRateLimitError, LLMAuthorizationError) as e:
+            raise CompletionRequestError(str(e))
+        except ValueError as e:
+            raise e
+        except Exception as e:
+            logging.exception("internal server error.")
+            raise InternalServerError()
+
+
+class CompletionStopApi(InstalledAppResource):
+    def post(self, installed_app, task_id):
+        app_model = installed_app.app
+        if app_model.mode != 'completion':
+            raise NotCompletionAppError()
+
+        PubHandler.stop(current_user, task_id)
+
+        return {'result': 'success'}, 200
+
+
+class ChatApi(InstalledAppResource):
+    def post(self, installed_app):
+        app_model = installed_app.app
+        if app_model.mode != 'chat':
+            raise NotChatAppError()
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('inputs', type=dict, required=True, location='json')
+        parser.add_argument('query', type=str, required=True, location='json')
+        parser.add_argument('response_mode', type=str, choices=['blocking', 'streaming'], location='json')
+        parser.add_argument('conversation_id', type=uuid_value, location='json')
+        args = parser.parse_args()
+
+        streaming = args['response_mode'] == 'streaming'
+
+        try:
+            response = CompletionService.completion(
+                app_model=app_model,
+                user=current_user,
+                args=args,
+                from_source='console',
+                streaming=streaming
+            )
+
+            return compact_response(response)
+        except services.errors.conversation.ConversationNotExistsError:
+            raise NotFound("Conversation Not Exists.")
+        except services.errors.conversation.ConversationCompletedError:
+            raise ConversationCompletedError()
+        except services.errors.app_model_config.AppModelConfigBrokenError:
+            logging.exception("App model config broken.")
+            raise AppUnavailableError()
+        except ProviderTokenNotInitError:
+            raise ProviderNotInitializeError()
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except (LLMBadRequestError, LLMAPIConnectionError, LLMAPIUnavailableError,
+                LLMRateLimitError, LLMAuthorizationError) as e:
+            raise CompletionRequestError(str(e))
+        except ValueError as e:
+            raise e
+        except Exception as e:
+            logging.exception("internal server error.")
+            raise InternalServerError()
+
+
+class ChatStopApi(InstalledAppResource):
+    def post(self, installed_app, task_id):
+        app_model = installed_app.app
+        if app_model.mode != 'chat':
+            raise NotChatAppError()
+
+        PubHandler.stop(current_user, task_id)
+
+        return {'result': 'success'}, 200
+
+
+def compact_response(response: Union[dict | Generator]) -> Response:
+    if isinstance(response, dict):
+        return Response(response=json.dumps(response), status=200, mimetype='application/json')
+    else:
+        def generate() -> Generator:
+            try:
+                for chunk in response:
+                    yield chunk
+            except services.errors.conversation.ConversationNotExistsError:
+                yield "data: " + json.dumps(api.handle_error(NotFound("Conversation Not Exists.")).get_json()) + "\n\n"
+            except services.errors.conversation.ConversationCompletedError:
+                yield "data: " + json.dumps(api.handle_error(ConversationCompletedError()).get_json()) + "\n\n"
+            except services.errors.app_model_config.AppModelConfigBrokenError:
+                logging.exception("App model config broken.")
+                yield "data: " + json.dumps(api.handle_error(AppUnavailableError()).get_json()) + "\n\n"
+            except ProviderTokenNotInitError:
+                yield "data: " + json.dumps(api.handle_error(ProviderNotInitializeError()).get_json()) + "\n\n"
+            except QuotaExceededError:
+                yield "data: " + json.dumps(api.handle_error(ProviderQuotaExceededError()).get_json()) + "\n\n"
+            except ModelCurrentlyNotSupportError:
+                yield "data: " + json.dumps(api.handle_error(ProviderModelCurrentlyNotSupportError()).get_json()) + "\n\n"
+            except (LLMBadRequestError, LLMAPIConnectionError, LLMAPIUnavailableError,
+                    LLMRateLimitError, LLMAuthorizationError) as e:
+                yield "data: " + json.dumps(api.handle_error(CompletionRequestError(str(e))).get_json()) + "\n\n"
+            except ValueError as e:
+                yield "data: " + json.dumps(api.handle_error(e).get_json()) + "\n\n"
+            except Exception:
+                logging.exception("internal server error.")
+                yield "data: " + json.dumps(api.handle_error(InternalServerError()).get_json()) + "\n\n"
+
+        return Response(stream_with_context(generate()), status=200,
+                        mimetype='text/event-stream')
+
+
+api.add_resource(CompletionApi, '/installed-apps/<uuid:installed_app_id>/completion-messages', endpoint='installed_app_completion')
+api.add_resource(CompletionStopApi, '/installed-apps/<uuid:installed_app_id>/completion-messages/<string:task_id>/stop', endpoint='installed_app_stop_completion')
+api.add_resource(ChatApi, '/installed-apps/<uuid:installed_app_id>/chat-messages', endpoint='installed_app_chat_completion')
+api.add_resource(ChatStopApi, '/installed-apps/<uuid:installed_app_id>/chat-messages/<string:task_id>/stop', endpoint='installed_app_stop_chat_completion')

api/controllers/console/explore/conversation.py

@@ -0,0 +1,127 @@
+# -*- coding:utf-8 -*-
+from flask_login import current_user
+from flask_restful import fields, reqparse, marshal_with
+from flask_restful.inputs import int_range
+from werkzeug.exceptions import NotFound
+
+from controllers.console import api
+from controllers.console.explore.error import NotChatAppError
+from controllers.console.explore.wraps import InstalledAppResource
+from libs.helper import TimestampField, uuid_value
+from services.conversation_service import ConversationService
+from services.errors.conversation import LastConversationNotExistsError, ConversationNotExistsError
+from services.web_conversation_service import WebConversationService
+
+conversation_fields = {
+    'id': fields.String,
+    'name': fields.String,
+    'inputs': fields.Raw,
+    'status': fields.String,
+    'introduction': fields.String,
+    'created_at': TimestampField
+}
+
+conversation_infinite_scroll_pagination_fields = {
+    'limit': fields.Integer,
+    'has_more': fields.Boolean,
+    'data': fields.List(fields.Nested(conversation_fields))
+}
+
+
+class ConversationListApi(InstalledAppResource):
+
+    @marshal_with(conversation_infinite_scroll_pagination_fields)
+    def get(self, installed_app):
+        app_model = installed_app.app
+        if app_model.mode != 'chat':
+            raise NotChatAppError()
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('last_id', type=uuid_value, location='args')
+        parser.add_argument('limit', type=int_range(1, 100), required=False, default=20, location='args')
+        parser.add_argument('pinned', type=str, choices=['true', 'false', None], location='args')
+        args = parser.parse_args()
+
+        pinned = None
+        if 'pinned' in args and args['pinned'] is not None:
+            pinned = True if args['pinned'] == 'true' else False
+
+        try:
+            return WebConversationService.pagination_by_last_id(
+                app_model=app_model,
+                user=current_user,
+                last_id=args['last_id'],
+                limit=args['limit'],
+                pinned=pinned
+            )
+        except LastConversationNotExistsError:
+            raise NotFound("Last Conversation Not Exists.")
+
+
+class ConversationApi(InstalledAppResource):
+    def delete(self, installed_app, c_id):
+        app_model = installed_app.app
+        if app_model.mode != 'chat':
+            raise NotChatAppError()
+
+        conversation_id = str(c_id)
+        ConversationService.delete(app_model, conversation_id, current_user)
+        WebConversationService.unpin(app_model, conversation_id, current_user)
+
+        return {"result": "success"}, 204
+
+
+class ConversationRenameApi(InstalledAppResource):
+
+    @marshal_with(conversation_fields)
+    def post(self, installed_app, c_id):
+        app_model = installed_app.app
+        if app_model.mode != 'chat':
+            raise NotChatAppError()
+
+        conversation_id = str(c_id)
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('name', type=str, required=True, location='json')
+        args = parser.parse_args()
+
+        try:
+            return ConversationService.rename(app_model, conversation_id, current_user, args['name'])
+        except ConversationNotExistsError:
+            raise NotFound("Conversation Not Exists.")
+
+
+class ConversationPinApi(InstalledAppResource):
+
+    def patch(self, installed_app, c_id):
+        app_model = installed_app.app
+        if app_model.mode != 'chat':
+            raise NotChatAppError()
+
+        conversation_id = str(c_id)
+
+        try:
+            WebConversationService.pin(app_model, conversation_id, current_user)
+        except ConversationNotExistsError:
+            raise NotFound("Conversation Not Exists.")
+
+        return {"result": "success"}
+
+
+class ConversationUnPinApi(InstalledAppResource):
+    def patch(self, installed_app, c_id):
+        app_model = installed_app.app
+        if app_model.mode != 'chat':
+            raise NotChatAppError()
+
+        conversation_id = str(c_id)
+        WebConversationService.unpin(app_model, conversation_id, current_user)
+
+        return {"result": "success"}
+
+
+api.add_resource(ConversationRenameApi, '/installed-apps/<uuid:installed_app_id>/conversations/<uuid:c_id>/name', endpoint='installed_app_conversation_rename')
+api.add_resource(ConversationListApi, '/installed-apps/<uuid:installed_app_id>/conversations', endpoint='installed_app_conversations')
+api.add_resource(ConversationApi, '/installed-apps/<uuid:installed_app_id>/conversations/<uuid:c_id>', endpoint='installed_app_conversation')
+api.add_resource(ConversationPinApi, '/installed-apps/<uuid:installed_app_id>/conversations/<uuid:c_id>/pin', endpoint='installed_app_conversation_pin')
+api.add_resource(ConversationUnPinApi, '/installed-apps/<uuid:installed_app_id>/conversations/<uuid:c_id>/unpin', endpoint='installed_app_conversation_unpin')

api/controllers/console/explore/error.py

@@ -0,0 +1,20 @@
+# -*- coding:utf-8 -*-
+from libs.exception import BaseHTTPException
+
+
+class NotCompletionAppError(BaseHTTPException):
+    error_code = 'not_completion_app'
+    description = "Not Completion App"
+    code = 400
+
+
+class NotChatAppError(BaseHTTPException):
+    error_code = 'not_chat_app'
+    description = "Not Chat App"
+    code = 400
+
+
+class AppSuggestedQuestionsAfterAnswerDisabledError(BaseHTTPException):
+    error_code = 'app_suggested_questions_after_answer_disabled'
+    description = "Function Suggested questions after answer disabled."
+    code = 403

api/controllers/console/explore/installed_app.py

@@ -0,0 +1,143 @@
+# -*- coding:utf-8 -*-
+from datetime import datetime
+
+from flask_login import login_required, current_user
+from flask_restful import Resource, reqparse, fields, marshal_with, inputs
+from sqlalchemy import and_
+from werkzeug.exceptions import NotFound, Forbidden, BadRequest
+
+from controllers.console import api
+from controllers.console.explore.wraps import InstalledAppResource
+from controllers.console.wraps import account_initialization_required
+from extensions.ext_database import db
+from libs.helper import TimestampField
+from models.model import App, InstalledApp, RecommendedApp
+from services.account_service import TenantService
+
+app_fields = {
+    'id': fields.String,
+    'name': fields.String,
+    'mode': fields.String,
+    'icon': fields.String,
+    'icon_background': fields.String
+}
+
+installed_app_fields = {
+    'id': fields.String,
+    'app': fields.Nested(app_fields),
+    'app_owner_tenant_id': fields.String,
+    'is_pinned': fields.Boolean,
+    'last_used_at': TimestampField,
+    'editable': fields.Boolean,
+    'uninstallable': fields.Boolean,
+}
+
+installed_app_list_fields = {
+    'installed_apps': fields.List(fields.Nested(installed_app_fields))
+}
+
+
+class InstalledAppsListApi(Resource):
+    @login_required
+    @account_initialization_required
+    @marshal_with(installed_app_list_fields)
+    def get(self):
+        current_tenant_id = current_user.current_tenant_id
+        installed_apps = db.session.query(InstalledApp).filter(
+            InstalledApp.tenant_id == current_tenant_id
+        ).all()
+
+        current_user.role = TenantService.get_user_role(current_user, current_user.current_tenant)
+        installed_apps = [
+            {
+                'id': installed_app.id,
+                'app': installed_app.app,
+                'app_owner_tenant_id': installed_app.app_owner_tenant_id,
+                'is_pinned': installed_app.is_pinned,
+                'last_used_at': installed_app.last_used_at,
+                "editable": current_user.role in ["owner", "admin"],
+                "uninstallable": current_tenant_id == installed_app.app_owner_tenant_id
+            }
+            for installed_app in installed_apps
+        ]
+        installed_apps.sort(key=lambda app: (-app['is_pinned'], app['last_used_at']
+                            if app['last_used_at'] is not None else datetime.min))
+
+        return {'installed_apps': installed_apps}
+
+    @login_required
+    @account_initialization_required
+    def post(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument('app_id', type=str, required=True, help='Invalid app_id')
+        args = parser.parse_args()
+
+        recommended_app = RecommendedApp.query.filter(RecommendedApp.app_id == args['app_id']).first()
+        if recommended_app is None:
+            raise NotFound('App not found')
+
+        current_tenant_id = current_user.current_tenant_id
+        app = db.session.query(App).filter(
+            App.id == args['app_id']
+        ).first()
+
+        if app is None:
+            raise NotFound('App not found')
+
+        if not app.is_public:
+            raise Forbidden('You can\'t install a non-public app')
+
+        installed_app = InstalledApp.query.filter(and_(
+            InstalledApp.app_id == args['app_id'],
+            InstalledApp.tenant_id == current_tenant_id
+        )).first()
+
+        if installed_app is None:
+            # todo: position
+            recommended_app.install_count += 1
+
+            new_installed_app = InstalledApp(
+                app_id=args['app_id'],
+                tenant_id=current_tenant_id,
+                app_owner_tenant_id=app.tenant_id,
+                is_pinned=False,
+                last_used_at=datetime.utcnow()
+            )
+            db.session.add(new_installed_app)
+            db.session.commit()
+
+        return {'message': 'App installed successfully'}
+
+
+class InstalledAppApi(InstalledAppResource):
+    """
+    update and delete an installed app
+    use InstalledAppResource to apply default decorators and get installed_app
+    """
+    def delete(self, installed_app):
+        if installed_app.app_owner_tenant_id == current_user.current_tenant_id:
+            raise BadRequest('You can\'t uninstall an app owned by the current tenant')
+
+        db.session.delete(installed_app)
+        db.session.commit()
+
+        return {'result': 'success', 'message': 'App uninstalled successfully'}
+
+    def patch(self, installed_app):
+        parser = reqparse.RequestParser()
+        parser.add_argument('is_pinned', type=inputs.boolean)
+        args = parser.parse_args()
+
+        commit_args = False
+        if 'is_pinned' in args:
+            installed_app.is_pinned = args['is_pinned']
+            commit_args = True
+
+        if commit_args:
+            db.session.commit()
+
+        return {'result': 'success', 'message': 'App info updated successfully'}
+
+
+api.add_resource(InstalledAppsListApi, '/installed-apps')
+api.add_resource(InstalledAppApi, '/installed-apps/<uuid:installed_app_id>')

api/controllers/console/explore/message.py

@@ -0,0 +1,196 @@
+# -*- coding:utf-8 -*-
+import json
+import logging
+from typing import Generator, Union
+
+from flask import stream_with_context, Response
+from flask_login import current_user
+from flask_restful import reqparse, fields, marshal_with
+from flask_restful.inputs import int_range
+from werkzeug.exceptions import NotFound, InternalServerError
+
+import services
+from controllers.console import api
+from controllers.console.app.error import AppMoreLikeThisDisabledError, ProviderNotInitializeError, \
+    ProviderQuotaExceededError, ProviderModelCurrentlyNotSupportError, CompletionRequestError
+from controllers.console.explore.error import NotCompletionAppError, AppSuggestedQuestionsAfterAnswerDisabledError
+from controllers.console.explore.wraps import InstalledAppResource
+from core.llm.error import LLMRateLimitError, LLMBadRequestError, LLMAuthorizationError, LLMAPIConnectionError, \
+    ProviderTokenNotInitError, LLMAPIUnavailableError, QuotaExceededError, ModelCurrentlyNotSupportError
+from libs.helper import uuid_value, TimestampField
+from services.completion_service import CompletionService
+from services.errors.app import MoreLikeThisDisabledError
+from services.errors.conversation import ConversationNotExistsError
+from services.errors.message import MessageNotExistsError, SuggestedQuestionsAfterAnswerDisabledError
+from services.message_service import MessageService
+
+
+class MessageListApi(InstalledAppResource):
+    feedback_fields = {
+        'rating': fields.String
+    }
+
+    message_fields = {
+        'id': fields.String,
+        'conversation_id': fields.String,
+        'inputs': fields.Raw,
+        'query': fields.String,
+        'answer': fields.String,
+        'feedback': fields.Nested(feedback_fields, attribute='user_feedback', allow_null=True),
+        'created_at': TimestampField
+    }
+
+    message_infinite_scroll_pagination_fields = {
+        'limit': fields.Integer,
+        'has_more': fields.Boolean,
+        'data': fields.List(fields.Nested(message_fields))
+    }
+
+    @marshal_with(message_infinite_scroll_pagination_fields)
+    def get(self, installed_app):
+        app_model = installed_app.app
+
+        if app_model.mode != 'chat':
+            raise NotChatAppError()
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('conversation_id', required=True, type=uuid_value, location='args')
+        parser.add_argument('first_id', type=uuid_value, location='args')
+        parser.add_argument('limit', type=int_range(1, 100), required=False, default=20, location='args')
+        args = parser.parse_args()
+
+        try:
+            return MessageService.pagination_by_first_id(app_model, current_user,
+                                                     args['conversation_id'], args['first_id'], args['limit'])
+        except services.errors.conversation.ConversationNotExistsError:
+            raise NotFound("Conversation Not Exists.")
+        except services.errors.message.FirstMessageNotExistsError:
+            raise NotFound("First Message Not Exists.")
+
+
+class MessageFeedbackApi(InstalledAppResource):
+    def post(self, installed_app, message_id):
+        app_model = installed_app.app
+
+        message_id = str(message_id)
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('rating', type=str, choices=['like', 'dislike', None], location='json')
+        args = parser.parse_args()
+
+        try:
+            MessageService.create_feedback(app_model, message_id, current_user, args['rating'])
+        except services.errors.message.MessageNotExistsError:
+            raise NotFound("Message Not Exists.")
+
+        return {'result': 'success'}
+
+
+class MessageMoreLikeThisApi(InstalledAppResource):
+    def get(self, installed_app, message_id):
+        app_model = installed_app.app
+        if app_model.mode != 'completion':
+            raise NotCompletionAppError()
+
+        message_id = str(message_id)
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('response_mode', type=str, required=True, choices=['blocking', 'streaming'], location='args')
+        args = parser.parse_args()
+
+        streaming = args['response_mode'] == 'streaming'
+
+        try:
+            response = CompletionService.generate_more_like_this(app_model, current_user, message_id, streaming)
+            return compact_response(response)
+        except MessageNotExistsError:
+            raise NotFound("Message Not Exists.")
+        except MoreLikeThisDisabledError:
+            raise AppMoreLikeThisDisabledError()
+        except ProviderTokenNotInitError:
+            raise ProviderNotInitializeError()
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except (LLMBadRequestError, LLMAPIConnectionError, LLMAPIUnavailableError,
+                LLMRateLimitError, LLMAuthorizationError) as e:
+            raise CompletionRequestError(str(e))
+        except ValueError as e:
+            raise e
+        except Exception:
+            logging.exception("internal server error.")
+            raise InternalServerError()
+
+
+def compact_response(response: Union[dict | Generator]) -> Response:
+    if isinstance(response, dict):
+        return Response(response=json.dumps(response), status=200, mimetype='application/json')
+    else:
+        def generate() -> Generator:
+            try:
+                for chunk in response:
+                    yield chunk
+            except MessageNotExistsError:
+                yield "data: " + json.dumps(api.handle_error(NotFound("Message Not Exists.")).get_json()) + "\n\n"
+            except MoreLikeThisDisabledError:
+                yield "data: " + json.dumps(api.handle_error(AppMoreLikeThisDisabledError()).get_json()) + "\n\n"
+            except ProviderTokenNotInitError:
+                yield "data: " + json.dumps(api.handle_error(ProviderNotInitializeError()).get_json()) + "\n\n"
+            except QuotaExceededError:
+                yield "data: " + json.dumps(api.handle_error(ProviderQuotaExceededError()).get_json()) + "\n\n"
+            except ModelCurrentlyNotSupportError:
+                yield "data: " + json.dumps(api.handle_error(ProviderModelCurrentlyNotSupportError()).get_json()) + "\n\n"
+            except (LLMBadRequestError, LLMAPIConnectionError, LLMAPIUnavailableError,
+                    LLMRateLimitError, LLMAuthorizationError) as e:
+                yield "data: " + json.dumps(api.handle_error(CompletionRequestError(str(e))).get_json()) + "\n\n"
+            except ValueError as e:
+                yield "data: " + json.dumps(api.handle_error(e).get_json()) + "\n\n"
+            except Exception:
+                logging.exception("internal server error.")
+                yield "data: " + json.dumps(api.handle_error(InternalServerError()).get_json()) + "\n\n"
+
+        return Response(stream_with_context(generate()), status=200,
+                        mimetype='text/event-stream')
+
+
+class MessageSuggestedQuestionApi(InstalledAppResource):
+    def get(self, installed_app, message_id):
+        app_model = installed_app.app
+        if app_model.mode != 'chat':
+            raise NotCompletionAppError()
+
+        message_id = str(message_id)
+
+        try:
+            questions = MessageService.get_suggested_questions_after_answer(
+                app_model=app_model,
+                user=current_user,
+                message_id=message_id
+            )
+        except MessageNotExistsError:
+            raise NotFound("Message not found")
+        except ConversationNotExistsError:
+            raise NotFound("Conversation not found")
+        except SuggestedQuestionsAfterAnswerDisabledError:
+            raise AppSuggestedQuestionsAfterAnswerDisabledError()
+        except ProviderTokenNotInitError:
+            raise ProviderNotInitializeError()
+        except QuotaExceededError:
+            raise ProviderQuotaExceededError()
+        except ModelCurrentlyNotSupportError:
+            raise ProviderModelCurrentlyNotSupportError()
+        except (LLMBadRequestError, LLMAPIConnectionError, LLMAPIUnavailableError,
+                LLMRateLimitError, LLMAuthorizationError) as e:
+            raise CompletionRequestError(str(e))
+        except Exception:
+            logging.exception("internal server error.")
+            raise InternalServerError()
+
+        return {'data': questions}
+
+
+api.add_resource(MessageListApi, '/installed-apps/<uuid:installed_app_id>/messages', endpoint='installed_app_messages')
+api.add_resource(MessageFeedbackApi, '/installed-apps/<uuid:installed_app_id>/messages/<uuid:message_id>/feedbacks', endpoint='installed_app_message_feedback')
+api.add_resource(MessageMoreLikeThisApi, '/installed-apps/<uuid:installed_app_id>/messages/<uuid:message_id>/more-like-this', endpoint='installed_app_more_like_this')
+api.add_resource(MessageSuggestedQuestionApi, '/installed-apps/<uuid:installed_app_id>/messages/<uuid:message_id>/suggested-questions', endpoint='installed_app_suggested_question')

api/controllers/console/explore/parameter.py

@@ -0,0 +1,43 @@
+# -*- coding:utf-8 -*-
+from flask_restful import marshal_with, fields
+
+from controllers.console import api
+from controllers.console.explore.wraps import InstalledAppResource
+
+
+class AppParameterApi(InstalledAppResource):
+    """Resource for app variables."""
+    variable_fields = {
+        'key': fields.String,
+        'name': fields.String,
+        'description': fields.String,
+        'type': fields.String,
+        'default': fields.String,
+        'max_length': fields.Integer,
+        'options': fields.List(fields.String)
+    }
+
+    parameters_fields = {
+        'opening_statement': fields.String,
+        'suggested_questions': fields.Raw,
+        'suggested_questions_after_answer': fields.Raw,
+        'more_like_this': fields.Raw,
+        'user_input_form': fields.Raw,
+    }
+
+    @marshal_with(parameters_fields)
+    def get(self, installed_app):
+        """Retrieve app parameters."""
+        app_model = installed_app.app
+        app_model_config = app_model.app_model_config
+
+        return {
+            'opening_statement': app_model_config.opening_statement,
+            'suggested_questions': app_model_config.suggested_questions_list,
+            'suggested_questions_after_answer': app_model_config.suggested_questions_after_answer_dict,
+            'more_like_this': app_model_config.more_like_this_dict,
+            'user_input_form': app_model_config.user_input_form_list
+        }
+
+
+api.add_resource(AppParameterApi, '/installed-apps/<uuid:installed_app_id>/parameters', endpoint='installed_app_parameters')

api/controllers/console/explore/recommended_app.py

@@ -0,0 +1,138 @@
+# -*- coding:utf-8 -*-
+from flask_login import login_required, current_user
+from flask_restful import Resource, fields, marshal_with
+from sqlalchemy import and_
+
+from controllers.console import api
+from controllers.console.app.error import AppNotFoundError
+from controllers.console.wraps import account_initialization_required
+from extensions.ext_database import db
+from models.model import App, InstalledApp, RecommendedApp
+from services.account_service import TenantService
+
+app_fields = {
+    'id': fields.String,
+    'name': fields.String,
+    'mode': fields.String,
+    'icon': fields.String,
+    'icon_background': fields.String
+}
+
+recommended_app_fields = {
+    'app': fields.Nested(app_fields, attribute='app'),
+    'app_id': fields.String,
+    'description': fields.String(attribute='description'),
+    'copyright': fields.String,
+    'privacy_policy': fields.String,
+    'category': fields.String,
+    'position': fields.Integer,
+    'is_listed': fields.Boolean,
+    'install_count': fields.Integer,
+    'installed': fields.Boolean,
+    'editable': fields.Boolean
+}
+
+recommended_app_list_fields = {
+    'recommended_apps': fields.List(fields.Nested(recommended_app_fields)),
+    'categories': fields.List(fields.String)
+}
+
+
+class RecommendedAppListApi(Resource):
+    @login_required
+    @account_initialization_required
+    @marshal_with(recommended_app_list_fields)
+    def get(self):
+        language_prefix = current_user.interface_language if current_user.interface_language else 'en-US'
+
+        recommended_apps = db.session.query(RecommendedApp).filter(
+            RecommendedApp.is_listed == True,
+            RecommendedApp.language == language_prefix
+        ).all()
+
+        categories = set()
+        current_user.role = TenantService.get_user_role(current_user, current_user.current_tenant)
+        recommended_apps_result = []
+        for recommended_app in recommended_apps:
+            installed = db.session.query(InstalledApp).filter(
+                and_(
+                    InstalledApp.app_id == recommended_app.app_id,
+                    InstalledApp.tenant_id == current_user.current_tenant_id
+                )
+            ).first() is not None
+
+            app = recommended_app.app
+            if not app or not app.is_public:
+                continue
+
+            site = app.site
+            if not site:
+                continue
+
+            recommended_app_result = {
+                'id': recommended_app.id,
+                'app': app,
+                'app_id': recommended_app.app_id,
+                'description': site.description,
+                'copyright': site.copyright,
+                'privacy_policy': site.privacy_policy,
+                'category': recommended_app.category,
+                'position': recommended_app.position,
+                'is_listed': recommended_app.is_listed,
+                'install_count': recommended_app.install_count,
+                'installed': installed,
+                'editable': current_user.role in ['owner', 'admin'],
+            }
+            recommended_apps_result.append(recommended_app_result)
+
+            categories.add(recommended_app.category)  # add category to categories
+
+        return {'recommended_apps': recommended_apps_result, 'categories': list(categories)}
+
+
+class RecommendedAppApi(Resource):
+    model_config_fields = {
+        'opening_statement': fields.String,
+        'suggested_questions': fields.Raw(attribute='suggested_questions_list'),
+        'suggested_questions_after_answer': fields.Raw(attribute='suggested_questions_after_answer_dict'),
+        'more_like_this': fields.Raw(attribute='more_like_this_dict'),
+        'model': fields.Raw(attribute='model_dict'),
+        'user_input_form': fields.Raw(attribute='user_input_form_list'),
+        'pre_prompt': fields.String,
+        'agent_mode': fields.Raw(attribute='agent_mode_dict'),
+    }
+
+    app_simple_detail_fields = {
+        'id': fields.String,
+        'name': fields.String,
+        'icon': fields.String,
+        'icon_background': fields.String,
+        'mode': fields.String,
+        'app_model_config': fields.Nested(model_config_fields),
+    }
+
+    @login_required
+    @account_initialization_required
+    @marshal_with(app_simple_detail_fields)
+    def get(self, app_id):
+        app_id = str(app_id)
+
+        # is in public recommended list
+        recommended_app = db.session.query(RecommendedApp).filter(
+            RecommendedApp.is_listed == True,
+            RecommendedApp.app_id == app_id
+        ).first()
+
+        if not recommended_app:
+            raise AppNotFoundError
+
+        # get app detail
+        app = db.session.query(App).filter(App.id == app_id).first()
+        if not app or not app.is_public:
+            raise AppNotFoundError
+
+        return app
+
+
+api.add_resource(RecommendedAppListApi, '/explore/apps')
+api.add_resource(RecommendedAppApi, '/explore/apps/<uuid:app_id>')

api/controllers/console/explore/saved_message.py

@@ -0,0 +1,79 @@
+from flask_login import current_user
+from flask_restful import reqparse, marshal_with, fields
+from flask_restful.inputs import int_range
+from werkzeug.exceptions import NotFound
+
+from controllers.console import api
+from controllers.console.explore.error import NotCompletionAppError
+from controllers.console.explore.wraps import InstalledAppResource
+from libs.helper import uuid_value, TimestampField
+from services.errors.message import MessageNotExistsError
+from services.saved_message_service import SavedMessageService
+
+feedback_fields = {
+    'rating': fields.String
+}
+
+message_fields = {
+    'id': fields.String,
+    'inputs': fields.Raw,
+    'query': fields.String,
+    'answer': fields.String,
+    'feedback': fields.Nested(feedback_fields, attribute='user_feedback', allow_null=True),
+    'created_at': TimestampField
+}
+
+
+class SavedMessageListApi(InstalledAppResource):
+    saved_message_infinite_scroll_pagination_fields = {
+        'limit': fields.Integer,
+        'has_more': fields.Boolean,
+        'data': fields.List(fields.Nested(message_fields))
+    }
+
+    @marshal_with(saved_message_infinite_scroll_pagination_fields)
+    def get(self, installed_app):
+        app_model = installed_app.app
+        if app_model.mode != 'completion':
+            raise NotCompletionAppError()
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('last_id', type=uuid_value, location='args')
+        parser.add_argument('limit', type=int_range(1, 100), required=False, default=20, location='args')
+        args = parser.parse_args()
+
+        return SavedMessageService.pagination_by_last_id(app_model, current_user, args['last_id'], args['limit'])
+
+    def post(self, installed_app):
+        app_model = installed_app.app
+        if app_model.mode != 'completion':
+            raise NotCompletionAppError()
+
+        parser = reqparse.RequestParser()
+        parser.add_argument('message_id', type=uuid_value, required=True, location='json')
+        args = parser.parse_args()
+
+        try:
+            SavedMessageService.save(app_model, current_user, args['message_id'])
+        except MessageNotExistsError:
+            raise NotFound("Message Not Exists.")
+
+        return {'result': 'success'}
+
+
+class SavedMessageApi(InstalledAppResource):
+    def delete(self, installed_app, message_id):
+        app_model = installed_app.app
+
+        message_id = str(message_id)
+
+        if app_model.mode != 'completion':
+            raise NotCompletionAppError()
+
+        SavedMessageService.delete(app_model, current_user, message_id)
+
+        return {'result': 'success'}
+
+
+api.add_resource(SavedMessageListApi, '/installed-apps/<uuid:installed_app_id>/saved-messages', endpoint='installed_app_saved_messages')
+api.add_resource(SavedMessageApi, '/installed-apps/<uuid:installed_app_id>/saved-messages/<uuid:message_id>', endpoint='installed_app_saved_message')

api/controllers/console/explore/wraps.py

@@ -0,0 +1,48 @@
+from flask_login import login_required, current_user
+from flask_restful import Resource
+from functools import wraps
+
+from werkzeug.exceptions import NotFound
+
+from controllers.console.wraps import account_initialization_required
+from extensions.ext_database import db
+from models.model import InstalledApp
+
+
+def installed_app_required(view=None):
+    def decorator(view):
+        @wraps(view)
+        def decorated(*args, **kwargs):
+            if not kwargs.get('installed_app_id'):
+                raise ValueError('missing installed_app_id in path parameters')
+
+            installed_app_id = kwargs.get('installed_app_id')
+            installed_app_id = str(installed_app_id)
+
+            del kwargs['installed_app_id']
+
+            installed_app = db.session.query(InstalledApp).filter(
+                InstalledApp.id == str(installed_app_id),
+                InstalledApp.tenant_id == current_user.current_tenant_id
+            ).first()
+
+            if installed_app is None:
+                raise NotFound('Installed app not found')
+
+            if not installed_app.app:
+                db.session.delete(installed_app)
+                db.session.commit()
+
+                raise NotFound('Installed app not found')
+
+            return view(installed_app, *args, **kwargs)
+        return decorated
+
+    if view:
+        return decorator(view)
+    return decorator
+
+
+class InstalledAppResource(Resource):
+    # must be reversed if there are multiple decorators
+    method_decorators = [installed_app_required, account_initialization_required, login_required]

api/controllers/console/version.py

@@ -19,6 +19,14 @@ class VersionApi(Resource):
         args = parser.parse_args()
         check_update_url = current_app.config['CHECK_UPDATE_URL']
 
+        if not check_update_url:
+            return {
+                'version': '0.0.0',
+                'release_date': '',
+                'release_notes': '',
+                'can_auto_update': False
+            }
+
         try:
             response = requests.get(check_update_url, {
                 'current_version': args.get('current_version')

api/controllers/console/workspace/error.py

@@ -21,11 +21,11 @@ class InvalidInvitationCodeError(BaseHTTPException):
 
 class AccountAlreadyInitedError(BaseHTTPException):
     error_code = 'account_already_inited'
-    description = "Account already inited."
+    description = "The account has been initialized. Please refresh the page."
     code = 400
 
 
 class AccountNotInitializedError(BaseHTTPException):
     error_code = 'account_not_initialized'
-    description = "Account not initialized."
+    description = "The account has not been initialized yet. Please proceed with the initialization process first."
     code = 400

api/controllers/console/workspace/providers.py

@@ -82,29 +82,33 @@ class ProviderTokenApi(Resource):
 
         args = parser.parse_args()
 
-        if not args['token']:
-            raise ValueError('Token is empty')
+        if args['token']:
+            try:
+                ProviderService.validate_provider_configs(
+                    tenant=current_user.current_tenant,
+                    provider_name=ProviderName(provider),
+                    configs=args['token']
+                )
+                token_is_valid = True
+            except ValidateFailedError as ex:
+                raise ValueError(str(ex))
 
-        try:
-            ProviderService.validate_provider_configs(
+            base64_encrypted_token = ProviderService.get_encrypted_token(
                 tenant=current_user.current_tenant,
                 provider_name=ProviderName(provider),
                 configs=args['token']
             )
-            token_is_valid = True
-        except ValidateFailedError:
+        else:
+            base64_encrypted_token = None
             token_is_valid = False
 
         tenant = current_user.current_tenant
 
-        base64_encrypted_token = ProviderService.get_encrypted_token(
-            tenant=current_user.current_tenant,
-            provider_name=ProviderName(provider),
-            configs=args['token']
-        )
-
-        provider_model = Provider.query.filter_by(tenant_id=tenant.id, provider_name=provider,
-                                                  provider_type=ProviderType.CUSTOM.value).first()
+        provider_model = db.session.query(Provider).filter(
+                Provider.tenant_id == tenant.id,
+                Provider.provider_name == provider,
+                Provider.provider_type == ProviderType.CUSTOM.value
+            ).first()
 
         # Only allow updating token for CUSTOM provider type
         if provider_model:
@@ -117,6 +121,16 @@ class ProviderTokenApi(Resource):
                                       is_valid=token_is_valid)
             db.session.add(provider_model)
 
+        if provider_model.is_valid:
+            other_providers = db.session.query(Provider).filter(
+                Provider.tenant_id == tenant.id,
+                Provider.provider_name != provider,
+                Provider.provider_type == ProviderType.CUSTOM.value
+            ).all()
+
+            for other_provider in other_providers:
+                other_provider.is_valid = False
+
         db.session.commit()
 
         if provider in [ProviderName.ANTHROPIC.value, ProviderName.AZURE_OPENAI.value, ProviderName.COHERE.value,
@@ -143,7 +157,7 @@ class ProviderTokenValidateApi(Resource):
         args = parser.parse_args()
 
         # todo: remove this when the provider is supported
-        if provider in [ProviderName.ANTHROPIC.value, ProviderName.AZURE_OPENAI.value, ProviderName.COHERE.value,
+        if provider in [ProviderName.ANTHROPIC.value, ProviderName.COHERE.value,
                         ProviderName.HUGGINGFACEHUB.value]:
             return {'result': 'success', 'warning': 'MOCK: This provider is not supported yet.'}
 

api/controllers/service_api/app/error.py

@@ -4,43 +4,45 @@ from libs.exception import BaseHTTPException
 
 class AppUnavailableError(BaseHTTPException):
     error_code = 'app_unavailable'
-    description = "App unavailable."
+    description = "App unavailable, please check your app configurations."
     code = 400
 
 
 class NotCompletionAppError(BaseHTTPException):
     error_code = 'not_completion_app'
-    description = "Not Completion App"
+    description = "Please check if your Completion app mode matches the right API route."
     code = 400
 
 
 class NotChatAppError(BaseHTTPException):
     error_code = 'not_chat_app'
-    description = "Not Chat App"
+    description = "Please check if your Chat app mode matches the right API route."
     code = 400
 
 
 class ConversationCompletedError(BaseHTTPException):
     error_code = 'conversation_completed'
-    description = "Conversation Completed."
+    description = "The conversation has ended. Please start a new conversation."
     code = 400
 
 
 class ProviderNotInitializeError(BaseHTTPException):
     error_code = 'provider_not_initialize'
-    description = "Provider Token not initialize."
+    description = "No valid model provider credentials found. " \
+                  "Please go to Settings -> Model Provider to complete your provider credentials."
     code = 400
 
 
 class ProviderQuotaExceededError(BaseHTTPException):
     error_code = 'provider_quota_exceeded'
-    description = "Provider quota exceeded."
+    description = "Your quota for Dify Hosted OpenAI has been exhausted. " \
+                  "Please go to Settings -> Model Provider to complete your own provider credentials."
     code = 400
 
 
 class ProviderModelCurrentlyNotSupportError(BaseHTTPException):
     error_code = 'model_currently_not_support'
-    description = "GPT-4 currently not support."
+    description = "Dify Hosted OpenAI trial currently not support the GPT-4 model."
     code = 400
 
 

api/controllers/service_api/dataset/document.py

@@ -69,12 +69,16 @@ class DocumentListApi(DatasetApiResource):
         document_data = {
             'data_source': {
                 'type': 'upload_file',
-                'info': upload_file.id
+                'info': [
+                    {
+                        'upload_file_id': upload_file.id
+                    }
+                ]
             }
         }
 
         try:
-            document = DocumentService.save_document_with_dataset_id(
+            documents, batch = DocumentService.save_document_with_dataset_id(
                 dataset=dataset,
                 document_data=document_data,
                 account=dataset.created_by_account,
@@ -83,7 +87,7 @@ class DocumentListApi(DatasetApiResource):
             )
         except ProviderTokenNotInitError:
             raise ProviderNotInitializeError()
-
+        document = documents[0]
         if doc_type and doc_metadata:
             metadata_schema = DocumentService.DOCUMENT_METADATA_SCHEMA[doc_type]
 

api/controllers/service_api/dataset/error.py

@@ -16,5 +16,5 @@ class DocumentIndexingError(BaseHTTPException):
 
 class DatasetNotInitedError(BaseHTTPException):
     error_code = 'dataset_not_inited'
-    description = "Dataset not inited."
+    description = "The dataset is still being initialized or indexing. Please wait a moment."
     code = 403

api/controllers/web/conversation.py

@@ -47,7 +47,7 @@ class ConversationListApi(WebApiResource):
         try:
             return WebConversationService.pagination_by_last_id(
                 app_model=app_model,
-                end_user=end_user,
+                user=end_user,
                 last_id=args['last_id'],
                 limit=args['limit'],
                 pinned=pinned

api/controllers/web/error.py

@@ -4,43 +4,45 @@ from libs.exception import BaseHTTPException
 
 class AppUnavailableError(BaseHTTPException):
     error_code = 'app_unavailable'
-    description = "App unavailable."
+    description = "App unavailable, please check your app configurations."
     code = 400
 
 
 class NotCompletionAppError(BaseHTTPException):
     error_code = 'not_completion_app'
-    description = "Not Completion App"
+    description = "Please check if your Completion app mode matches the right API route."
     code = 400
 
 
 class NotChatAppError(BaseHTTPException):
     error_code = 'not_chat_app'
-    description = "Not Chat App"
+    description = "Please check if your Chat app mode matches the right API route."
     code = 400
 
 
 class ConversationCompletedError(BaseHTTPException):
     error_code = 'conversation_completed'
-    description = "Conversation Completed."
+    description = "The conversation has ended. Please start a new conversation."
     code = 400
 
 
 class ProviderNotInitializeError(BaseHTTPException):
     error_code = 'provider_not_initialize'
-    description = "Provider Token not initialize."
+    description = "No valid model provider credentials found. " \
+                  "Please go to Settings -> Model Provider to complete your provider credentials."
     code = 400
 
 
 class ProviderQuotaExceededError(BaseHTTPException):
     error_code = 'provider_quota_exceeded'
-    description = "Provider quota exceeded."
+    description = "Your quota for Dify Hosted OpenAI has been exhausted. " \
+                  "Please go to Settings -> Model Provider to complete your own provider credentials."
     code = 400
 
 
 class ProviderModelCurrentlyNotSupportError(BaseHTTPException):
     error_code = 'model_currently_not_support'
-    description = "GPT-4 currently not support."
+    description = "Dify Hosted OpenAI trial currently not support the GPT-4 model."
     code = 400
 
 
@@ -52,11 +54,11 @@ class CompletionRequestError(BaseHTTPException):
 
 class AppMoreLikeThisDisabledError(BaseHTTPException):
     error_code = 'app_more_like_this_disabled'
-    description = "More like this disabled."
+    description = "The 'More like this' feature is disabled. Please refresh your page."
     code = 403
 
 
 class AppSuggestedQuestionsAfterAnswerDisabledError(BaseHTTPException):
     error_code = 'app_suggested_questions_after_answer_disabled'
-    description = "Function Suggested questions after answer disabled."
+    description = "The 'Suggested Questions After Answer' feature is disabled. Please refresh your page."
     code = 403

api/controllers/web/wraps.py

@@ -16,7 +16,7 @@ def validate_token(view=None):
         def decorated(*args, **kwargs):
             site = validate_and_get_site()
 
-            app_model = db.session.query(App).get(site.app_id)
+            app_model = db.session.query(App).filter(App.id == site.app_id).first()
             if not app_model:
                 raise NotFound()
 
@@ -42,13 +42,16 @@ def validate_and_get_site():
     """
     auth_header = request.headers.get('Authorization')
     if auth_header is None:
-        raise Unauthorized()
+        raise Unauthorized('Authorization header is missing.')
+
+    if ' ' not in auth_header:
+        raise Unauthorized('Invalid Authorization header format. Expected \'Bearer <api-key>\' format.')
 
     auth_scheme, auth_token = auth_header.split(None, 1)
     auth_scheme = auth_scheme.lower()
 
     if auth_scheme != 'bearer':
-        raise Unauthorized()
+        raise Unauthorized('Invalid Authorization header format. Expected \'Bearer <api-key>\' format.')
 
     site = db.session.query(Site).filter(
         Site.code == auth_token,

api/core/callback_handler/index_tool_callback_handler.py

@@ -34,5 +34,9 @@ class DatasetIndexToolCallbackHandler(IndexToolCallbackHandler):
             db.session.query(DocumentSegment).filter(
                 DocumentSegment.dataset_id == self.dataset_id,
                 DocumentSegment.index_node_id == index_node_id
-            ).update({DocumentSegment.hit_count: DocumentSegment.hit_count + 1}, synchronize_session=False)
+            ).update(
+                {DocumentSegment.hit_count: DocumentSegment.hit_count + 1},
+                synchronize_session=False
+            )
 
+            db.session.commit()

api/core/callback_handler/llm_callback_handler.py

@@ -75,7 +75,12 @@ class LLMCallbackHandler(BaseCallbackHandler):
         self.conversation_message_task.save_message(self.llm_message)
 
     def on_llm_new_token(self, token: str, **kwargs: Any) -> None:
-        self.conversation_message_task.append_message_text(token)
+        try:
+            self.conversation_message_task.append_message_text(token)
+        except ConversationTaskStoppedException as ex:
+            self.on_llm_error(error=ex)
+            raise ex
+
         self.llm_message.completion += token
 
     def on_llm_error(

api/core/chain/llm_router_chain.py

@@ -0,0 +1,109 @@
+"""Base classes for LLM-powered router chains."""
+from __future__ import annotations
+
+import json
+from typing import Any, Dict, List, Optional, Type, cast, NamedTuple
+
+from langchain.chains.base import Chain
+from pydantic import root_validator
+
+from langchain.chains import LLMChain
+from langchain.prompts import BasePromptTemplate
+from langchain.schema import BaseOutputParser, OutputParserException, BaseLanguageModel
+
+from libs.json_in_md_parser import parse_and_check_json_markdown
+
+
+class Route(NamedTuple):
+    destination: Optional[str]
+    next_inputs: Dict[str, Any]
+
+
+class LLMRouterChain(Chain):
+    """A router chain that uses an LLM chain to perform routing."""
+
+    llm_chain: LLMChain
+    """LLM chain used to perform routing"""
+
+    @root_validator()
+    def validate_prompt(cls, values: dict) -> dict:
+        prompt = values["llm_chain"].prompt
+        if prompt.output_parser is None:
+            raise ValueError(
+                "LLMRouterChain requires base llm_chain prompt to have an output"
+                " parser that converts LLM text output to a dictionary with keys"
+                " 'destination' and 'next_inputs'. Received a prompt with no output"
+                " parser."
+            )
+        return values
+
+    @property
+    def input_keys(self) -> List[str]:
+        """Will be whatever keys the LLM chain prompt expects.
+
+        :meta private:
+        """
+        return self.llm_chain.input_keys
+
+    def _validate_outputs(self, outputs: Dict[str, Any]) -> None:
+        super()._validate_outputs(outputs)
+        if not isinstance(outputs["next_inputs"], dict):
+            raise ValueError
+
+    def _call(
+        self,
+        inputs: Dict[str, Any]
+    ) -> Dict[str, Any]:
+        output = cast(
+            Dict[str, Any],
+            self.llm_chain.predict_and_parse(**inputs),
+        )
+        return output
+
+    @classmethod
+    def from_llm(
+        cls, llm: BaseLanguageModel, prompt: BasePromptTemplate, **kwargs: Any
+    ) -> LLMRouterChain:
+        """Convenience constructor."""
+        llm_chain = LLMChain(llm=llm, prompt=prompt)
+        return cls(llm_chain=llm_chain, **kwargs)
+
+    @property
+    def output_keys(self) -> List[str]:
+        return ["destination", "next_inputs"]
+
+    def route(self, inputs: Dict[str, Any]) -> Route:
+        result = self(inputs)
+        return Route(result["destination"], result["next_inputs"])
+
+
+class RouterOutputParser(BaseOutputParser[Dict[str, str]]):
+    """Parser for output of router chain int he multi-prompt chain."""
+
+    default_destination: str = "DEFAULT"
+    next_inputs_type: Type = str
+    next_inputs_inner_key: str = "input"
+
+    def parse(self, text: str) -> Dict[str, Any]:
+        try:
+            expected_keys = ["destination", "next_inputs"]
+            parsed = parse_and_check_json_markdown(text, expected_keys)
+            if not isinstance(parsed["destination"], str):
+                raise ValueError("Expected 'destination' to be a string.")
+            if not isinstance(parsed["next_inputs"], self.next_inputs_type):
+                raise ValueError(
+                    f"Expected 'next_inputs' to be {self.next_inputs_type}."
+                )
+            parsed["next_inputs"] = {self.next_inputs_inner_key: parsed["next_inputs"]}
+            if (
+                parsed["destination"].strip().lower()
+                == self.default_destination.lower()
+            ):
+                parsed["destination"] = None
+            else:
+                parsed["destination"] = parsed["destination"].strip()
+            return parsed
+        except Exception as e:
+            raise OutputParserException(
+                f"Parsing text\n{text}\n of llm router raised following error:\n{e}"
+            )

api/core/chain/main_chain_builder.py

@@ -1,18 +1,18 @@
 from typing import Optional, List
 
-from langchain.callbacks import SharedCallbackManager
+from langchain.callbacks import SharedCallbackManager, CallbackManager
 from langchain.chains import SequentialChain
 from langchain.chains.base import Chain
 from langchain.memory.chat_memory import BaseChatMemory
 
-from core.agent.agent_builder import AgentBuilder
 from core.callback_handler.agent_loop_gather_callback_handler import AgentLoopGatherCallbackHandler
-from core.callback_handler.dataset_tool_callback_handler import DatasetToolCallbackHandler
 from core.callback_handler.main_chain_gather_callback_handler import MainChainGatherCallbackHandler
+from core.callback_handler.std_out_callback_handler import DifyStdOutCallbackHandler
 from core.chain.chain_builder import ChainBuilder
-from core.constant import llm_constant
+from core.chain.multi_dataset_router_chain import MultiDatasetRouterChain
 from core.conversation_message_task import ConversationMessageTask
-from core.tool.dataset_tool_builder import DatasetToolBuilder
+from extensions.ext_database import db
+from models.dataset import Dataset
 
 
 class MainChainBuilder:
@@ -31,8 +31,7 @@ class MainChainBuilder:
             tenant_id=tenant_id,
             agent_mode=agent_mode,
             memory=memory,
-            dataset_tool_callback_handler=DatasetToolCallbackHandler(conversation_message_task),
-            agent_loop_gather_callback_handler=chain_callback_handler.agent_loop_gather_callback_handler
+            conversation_message_task=conversation_message_task
         )
         chains += tool_chains
 
@@ -59,15 +58,15 @@ class MainChainBuilder:
 
     @classmethod
     def get_agent_chains(cls, tenant_id: str, agent_mode: dict, memory: Optional[BaseChatMemory],
-                         dataset_tool_callback_handler: DatasetToolCallbackHandler,
-                         agent_loop_gather_callback_handler: AgentLoopGatherCallbackHandler):
+                         conversation_message_task: ConversationMessageTask):
         # agent mode
         chains = []
         if agent_mode and agent_mode.get('enabled'):
             tools = agent_mode.get('tools', [])
 
             pre_fixed_chains = []
-            agent_tools = []
+            # agent_tools = []
+            datasets = []
             for tool in tools:
                 tool_type = list(tool.keys())[0]
                 tool_config = list(tool.values())[0]
@@ -76,34 +75,27 @@ class MainChainBuilder:
                     if chain:
                         pre_fixed_chains.append(chain)
                 elif tool_type == "dataset":
-                    dataset_tool = DatasetToolBuilder.build_dataset_tool(
-                        tenant_id=tenant_id,
-                        dataset_id=tool_config.get("id"),
-                        response_mode='no_synthesizer',  # "compact"
-                        callback_handler=dataset_tool_callback_handler
-                    )
+                    # get dataset from dataset id
+                    dataset = db.session.query(Dataset).filter(
+                        Dataset.tenant_id == tenant_id,
+                        Dataset.id == tool_config.get("id")
+                    ).first()
 
-                    if dataset_tool:
-                        agent_tools.append(dataset_tool)
+                    if dataset:
+                        datasets.append(dataset)
 
             # add pre-fixed chains
             chains += pre_fixed_chains
 
-            if len(agent_tools) == 1:
+            if len(datasets) > 0:
                 # tool to chain
-                tool_chain = ChainBuilder.to_tool_chain(tool=agent_tools[0], output_key='tool_output')
-                chains.append(tool_chain)
-            elif len(agent_tools) > 1:
-                # build agent config
-                agent_chain = AgentBuilder.to_agent_chain(
+                multi_dataset_router_chain = MultiDatasetRouterChain.from_datasets(
                     tenant_id=tenant_id,
-                    tools=agent_tools,
-                    memory=memory,
-                    dataset_tool_callback_handler=dataset_tool_callback_handler,
-                    agent_loop_gather_callback_handler=agent_loop_gather_callback_handler
+                    datasets=datasets,
+                    conversation_message_task=conversation_message_task,
+                    callback_manager=CallbackManager([DifyStdOutCallbackHandler()])
                 )
-
-                chains.append(agent_chain)
+                chains.append(multi_dataset_router_chain)
 
         final_output_key = cls.get_chains_output_key(chains)
 

api/core/chain/multi_dataset_router_chain.py

@@ -0,0 +1,144 @@
+from typing import Mapping, List, Dict, Any, Optional
+
+from langchain import LLMChain, PromptTemplate, ConversationChain
+from langchain.callbacks import CallbackManager
+from langchain.chains.base import Chain
+from langchain.schema import BaseLanguageModel
+from pydantic import Extra
+
+from core.callback_handler.dataset_tool_callback_handler import DatasetToolCallbackHandler
+from core.callback_handler.std_out_callback_handler import DifyStdOutCallbackHandler
+from core.chain.llm_router_chain import LLMRouterChain, RouterOutputParser
+from core.conversation_message_task import ConversationMessageTask
+from core.llm.llm_builder import LLMBuilder
+from core.tool.dataset_tool_builder import DatasetToolBuilder
+from core.tool.llama_index_tool import EnhanceLlamaIndexTool
+from models.dataset import Dataset
+
+MULTI_PROMPT_ROUTER_TEMPLATE = """
+Given a raw text input to a language model select the model prompt best suited for \
+the input. You will be given the names of the available prompts and a description of \
+what the prompt is best suited for. You may also revise the original input if you \
+think that revising it will ultimately lead to a better response from the language \
+model.
+
+<< FORMATTING >>
+Return a markdown code snippet with a JSON object formatted to look like, \
+no any other string out of markdown code snippet:
+```json
+{{{{
+    "destination": string \\ name of the prompt to use or "DEFAULT"
+    "next_inputs": string \\ a potentially modified version of the original input
+}}}}
+```
+
+REMEMBER: "destination" MUST be one of the candidate prompt names specified below OR \
+it can be "DEFAULT" if the input is not well suited for any of the candidate prompts.
+REMEMBER: "next_inputs" can just be the original input if you don't think any \
+modifications are needed.
+
+<< CANDIDATE PROMPTS >>
+{destinations}
+
+<< INPUT >>
+{{input}}
+
+<< OUTPUT >>
+"""
+
+
+class MultiDatasetRouterChain(Chain):
+    """Use a single chain to route an input to one of multiple candidate chains."""
+
+    router_chain: LLMRouterChain
+    """Chain for deciding a destination chain and the input to it."""
+    dataset_tools: Mapping[str, EnhanceLlamaIndexTool]
+    """Map of name to candidate chains that inputs can be routed to."""
+
+    class Config:
+        """Configuration for this pydantic object."""
+
+        extra = Extra.forbid
+        arbitrary_types_allowed = True
+
+    @property
+    def input_keys(self) -> List[str]:
+        """Will be whatever keys the router chain prompt expects.
+
+        :meta private:
+        """
+        return self.router_chain.input_keys
+
+    @property
+    def output_keys(self) -> List[str]:
+        return ["text"]
+
+    @classmethod
+    def from_datasets(
+            cls,
+            tenant_id: str,
+            datasets: List[Dataset],
+            conversation_message_task: ConversationMessageTask,
+            **kwargs: Any,
+    ):
+        """Convenience constructor for instantiating from destination prompts."""
+        llm_callback_manager = CallbackManager([DifyStdOutCallbackHandler()])
+        llm = LLMBuilder.to_llm(
+            tenant_id=tenant_id,
+            model_name='gpt-3.5-turbo',
+            temperature=0,
+            max_tokens=1024,
+            callback_manager=llm_callback_manager
+        )
+
+        destinations = ["{}: {}".format(d.id, d.description.replace('\n', ' ') if d.description
+                        else ('useful for when you want to answer queries about the ' + d.name))
+                        for d in datasets]
+        destinations_str = "\n".join(destinations)
+        router_template = MULTI_PROMPT_ROUTER_TEMPLATE.format(
+            destinations=destinations_str
+        )
+        router_prompt = PromptTemplate(
+            template=router_template,
+            input_variables=["input"],
+            output_parser=RouterOutputParser(),
+        )
+        router_chain = LLMRouterChain.from_llm(llm, router_prompt)
+        dataset_tools = {}
+        for dataset in datasets:
+            dataset_tool = DatasetToolBuilder.build_dataset_tool(
+                dataset=dataset,
+                response_mode='no_synthesizer',  # "compact"
+                callback_handler=DatasetToolCallbackHandler(conversation_message_task)
+            )
+
+            if dataset_tool:
+                dataset_tools[dataset.id] = dataset_tool
+
+        return cls(
+            router_chain=router_chain,
+            dataset_tools=dataset_tools,
+            **kwargs,
+        )
+
+    def _call(
+        self,
+        inputs: Dict[str, Any]
+    ) -> Dict[str, Any]:
+        if len(self.dataset_tools) == 0:
+            return {"text": ''}
+        elif len(self.dataset_tools) == 1:
+            return {"text": next(iter(self.dataset_tools.values())).run(inputs['input'])}
+
+        route = self.router_chain.route(inputs)
+
+        if not route.destination:
+            return {"text": ''}
+        elif route.destination in self.dataset_tools:
+            return {"text": self.dataset_tools[route.destination].run(
+                route.next_inputs['input']
+            )}
+        else:
+            raise ValueError(
+                f"Received invalid destination chain name '{route.destination}'"
+            )

api/core/completion.py

@@ -1,14 +1,17 @@
-from typing import Optional, List, Union
+import logging
+from typing import Optional, List, Union, Tuple
 
 from langchain.callbacks import CallbackManager
 from langchain.chat_models.base import BaseChatModel
 from langchain.llms import BaseLLM
 from langchain.schema import BaseMessage, BaseLanguageModel, HumanMessage
+from requests.exceptions import ChunkedEncodingError
+
 from core.constant import llm_constant
 from core.callback_handler.llm_callback_handler import LLMCallbackHandler
 from core.callback_handler.std_out_callback_handler import DifyStreamingStdOutCallbackHandler, \
     DifyStdOutCallbackHandler
-from core.conversation_message_task import ConversationMessageTask, ConversationTaskStoppedException
+from core.conversation_message_task import ConversationMessageTask, ConversationTaskStoppedException, PubHandler
 from core.llm.error import LLMBadRequestError
 from core.llm.llm_builder import LLMBuilder
 from core.chain.main_chain_builder import MainChainBuilder
@@ -39,7 +42,8 @@ class Completion:
             memory = cls.get_memory_from_conversation(
                 tenant_id=app.tenant_id,
                 app_model_config=app_model_config,
-                conversation=conversation
+                conversation=conversation,
+                return_messages=False
             )
 
             inputs = conversation.inputs
@@ -83,6 +87,11 @@ class Completion:
             )
         except ConversationTaskStoppedException:
             return
+        except ChunkedEncodingError as e:
+            # Interrupt by LLM (like OpenAI), handle it.
+            logging.warning(f'ChunkedEncodingError: {e}')
+            conversation_message_task.end()
+            return
 
     @classmethod
     def run_final_llm(cls, tenant_id: str, mode: str, app_model_config: AppModelConfig, query: str, inputs: dict,
@@ -96,7 +105,7 @@ class Completion:
         )
 
         # get llm prompt
-        prompt = cls.get_main_llm_prompt(
+        prompt, stop_words = cls.get_main_llm_prompt(
             mode=mode,
             llm=final_llm,
             pre_prompt=app_model_config.pre_prompt,
@@ -114,30 +123,47 @@ class Completion:
             mode=mode
         )
 
-        response = final_llm.generate([prompt])
+        response = final_llm.generate([prompt], stop_words)
 
         return response
 
     @classmethod
-    def get_main_llm_prompt(cls, mode: str, llm: BaseLanguageModel, pre_prompt: str, query: str, inputs: dict, chain_output: Optional[str],
+    def get_main_llm_prompt(cls, mode: str, llm: BaseLanguageModel, pre_prompt: str, query: str, inputs: dict,
+                            chain_output: Optional[str],
                             memory: Optional[ReadOnlyConversationTokenDBBufferSharedMemory]) -> \
-            Union[str | List[BaseMessage]]:
+            Tuple[Union[str | List[BaseMessage]], Optional[List[str]]]:
+        # disable template string in query
+        query_params = OutLinePromptTemplate.from_template(template=query).input_variables
+        if query_params:
+            for query_param in query_params:
+                if query_param not in inputs:
+                    inputs[query_param] = '{' + query_param + '}'
+
         pre_prompt = PromptBuilder.process_template(pre_prompt) if pre_prompt else pre_prompt
         if mode == 'completion':
             prompt_template = OutLinePromptTemplate.from_template(
-                template=("Use the following pieces of [CONTEXT] to answer the question at the end. "
-                          "If you don't know the answer, "
-                          "just say that you don't know, don't try to make up an answer. \n"
-                          "```\n"
-                          "[CONTEXT]\n"
-                          "{context}\n"
-                          "```\n" if chain_output else "")
+                template=("""Use the following CONTEXT as your learned knowledge:
+[CONTEXT]
+{context}
+[END CONTEXT]
+
+When answer to user:
+- If you don't know, just say that you don't know.
+- If you don't know when you are not sure, ask for clarification. 
+Avoid mentioning that you obtained the information from the context.
+And answer according to the language of the user's question.
+""" if chain_output else "")
                          + (pre_prompt + "\n" if pre_prompt else "")
                          + "{query}\n"
             )
 
             if chain_output:
                 inputs['context'] = chain_output
+                context_params = OutLinePromptTemplate.from_template(template=chain_output).input_variables
+                if context_params:
+                    for context_param in context_params:
+                        if context_param not in inputs:
+                            inputs[context_param] = '{' + context_param + '}'
 
             prompt_inputs = {k: inputs[k] for k in prompt_template.input_variables if k in inputs}
             prompt_content = prompt_template.format(
@@ -147,64 +173,83 @@ class Completion:
 
             if isinstance(llm, BaseChatModel):
                 # use chat llm as completion model
-                return [HumanMessage(content=prompt_content)]
+                return [HumanMessage(content=prompt_content)], None
             else:
-                return prompt_content
+                return prompt_content, None
         else:
             messages: List[BaseMessage] = []
 
-            system_message = None
-            if pre_prompt:
-                # append pre prompt as system message
-                system_message = PromptBuilder.to_system_message(pre_prompt, inputs)
-
-            if chain_output:
-                # append context as system message, currently only use simple stuff prompt
-                context_message = PromptBuilder.to_system_message(
-                    """Use the following pieces of [CONTEXT] to answer the users question. 
-If you don't know the answer, just say that you don't know, don't try to make up an answer.
-```
-[CONTEXT]
-{context}
-```""",
-                    {'context': chain_output}
-                )
-
-                if not system_message:
-                    system_message = context_message
-                else:
-                    system_message.content = context_message.content + "\n\n" + system_message.content
-
-            if system_message:
-                messages.append(system_message)
-
             human_inputs = {
                 "query": query
             }
 
-            # construct main prompt
-            human_message = PromptBuilder.to_human_message(
-                prompt_content="{query}",
-                inputs=human_inputs
-            )
+            human_message_prompt = ""
+
+            if pre_prompt:
+                pre_prompt_inputs = {k: inputs[k] for k in
+                                     OutLinePromptTemplate.from_template(template=pre_prompt).input_variables
+                                     if k in inputs}
+
+                if pre_prompt_inputs:
+                    human_inputs.update(pre_prompt_inputs)
+
+            if chain_output:
+                human_inputs['context'] = chain_output
+                human_message_prompt += """Use the following CONTEXT as your learned knowledge.
+[CONTEXT]
+{context}
+[END CONTEXT]
+
+When answer to user:
+- If you don't know, just say that you don't know.
+- If you don't know when you are not sure, ask for clarification. 
+Avoid mentioning that you obtained the information from the context.
+And answer according to the language of the user's question.
+"""
+
+            if pre_prompt:
+                human_message_prompt += pre_prompt
+
+            query_prompt = "\nHuman: {query}\nAI: "
 
             if memory:
                 # append chat histories
-                tmp_messages = messages.copy() + [human_message]
-                curr_message_tokens = memory.llm.get_messages_tokens(tmp_messages)
-                rest_tokens = llm_constant.max_context_token_length[
-                                  memory.llm.model_name] - memory.llm.max_tokens - curr_message_tokens
+                tmp_human_message = PromptBuilder.to_human_message(
+                    prompt_content=human_message_prompt + query_prompt,
+                    inputs=human_inputs
+                )
+
+                curr_message_tokens = memory.llm.get_messages_tokens([tmp_human_message])
+                rest_tokens = llm_constant.max_context_token_length[memory.llm.model_name] \
+                              - memory.llm.max_tokens - curr_message_tokens
                 rest_tokens = max(rest_tokens, 0)
-                history_messages = cls.get_history_messages_from_memory(memory, rest_tokens)
-                messages += history_messages
+                histories = cls.get_history_messages_from_memory(memory, rest_tokens)
+
+                # disable template string in query
+                histories_params = OutLinePromptTemplate.from_template(template=histories).input_variables
+                if histories_params:
+                    for histories_param in histories_params:
+                        if histories_param not in human_inputs:
+                            human_inputs[histories_param] = '{' + histories_param + '}'
+
+                human_message_prompt += "\n\n" + histories
+
+            human_message_prompt += query_prompt
+
+            # construct main prompt
+            human_message = PromptBuilder.to_human_message(
+                prompt_content=human_message_prompt,
+                inputs=human_inputs
+            )
 
             messages.append(human_message)
 
-            return messages
+            return messages, ['\nHuman:']
 
     @classmethod
     def get_llm_callback_manager(cls, llm: Union[StreamableOpenAI, StreamableChatOpenAI],
-                                 streaming: bool, conversation_message_task: ConversationMessageTask) -> CallbackManager:
+                                 streaming: bool,
+                                 conversation_message_task: ConversationMessageTask) -> CallbackManager:
         llm_callback_handler = LLMCallbackHandler(llm, conversation_message_task)
         if streaming:
             callback_handlers = [llm_callback_handler, DifyStreamingStdOutCallbackHandler()]
@@ -216,7 +261,7 @@ If you don't know the answer, just say that you don't know, don't try to make up
     @classmethod
     def get_history_messages_from_memory(cls, memory: ReadOnlyConversationTokenDBBufferSharedMemory,
                                          max_token_limit: int) -> \
-            List[BaseMessage]:
+            str:
         """Get memory messages."""
         memory.max_token_limit = max_token_limit
         memory_key = memory.memory_variables[0]
@@ -286,7 +331,7 @@ If you don't know the answer, just say that you don't know, don't try to make up
         )
 
         # get llm prompt
-        original_prompt = cls.get_main_llm_prompt(
+        original_prompt, _ = cls.get_main_llm_prompt(
             mode="completion",
             llm=llm,
             pre_prompt=pre_prompt,

api/core/constant/llm_constant.py

@@ -4,6 +4,7 @@ models = {
     'gpt-4': 'openai',  # 8,192 tokens
     'gpt-4-32k': 'openai',  # 32,768 tokens
     'gpt-3.5-turbo': 'openai',  # 4,096 tokens
+    'gpt-3.5-turbo-16k': 'openai',  # 16384 tokens
     'text-davinci-003': 'openai',  # 4,097 tokens
     'text-davinci-002': 'openai',  # 4,097 tokens
     'text-curie-001': 'openai',  # 2,049 tokens
@@ -16,6 +17,7 @@ max_context_token_length = {
     'gpt-4': 8192,
     'gpt-4-32k': 32768,
     'gpt-3.5-turbo': 4096,
+    'gpt-3.5-turbo-16k': 16384,
     'text-davinci-003': 4097,
     'text-davinci-002': 4097,
     'text-curie-001': 2049,
@@ -29,11 +31,13 @@ models_by_mode = {
         'gpt-4',  # 8,192 tokens
         'gpt-4-32k',  # 32,768 tokens
         'gpt-3.5-turbo',  # 4,096 tokens
+        'gpt-3.5-turbo-16k',  # 16,384 tokens
     ],
     'completion': [
         'gpt-4',  # 8,192 tokens
         'gpt-4-32k',  # 32,768 tokens
         'gpt-3.5-turbo',  # 4,096 tokens
+        'gpt-3.5-turbo-16k',  # 16,384 tokens
         'text-davinci-003',  # 4,097 tokens
         'text-davinci-002'  # 4,097 tokens
         'text-curie-001',  # 2,049 tokens
@@ -57,9 +61,13 @@ model_prices = {
         'completion': Decimal('0.12')
     },
     'gpt-3.5-turbo': {
-        'prompt': Decimal('0.002'),
+        'prompt': Decimal('0.0015'),
         'completion': Decimal('0.002')
     },
+    'gpt-3.5-turbo-16k': {
+        'prompt': Decimal('0.003'),
+        'completion': Decimal('0.004')
+    },
     'text-davinci-003': {
         'prompt': Decimal('0.02'),
         'completion': Decimal('0.02')
@@ -77,7 +85,7 @@ model_prices = {
         'completion': Decimal('0.0004')
     },
     'text-embedding-ada-002': {
-        'usage': Decimal('0.0004'),
+        'usage': Decimal('0.0001'),
     }
 }
 

api/core/conversation_message_task.py

@@ -2,8 +2,6 @@ import decimal
 import json
 from typing import Optional, Union
 
-from gunicorn.config import User
-
 from core.callback_handler.entity.agent_loop import AgentLoop
 from core.callback_handler.entity.dataset_query import DatasetQueryObj
 from core.callback_handler.entity.llm_message import LLMMessage
@@ -58,6 +56,9 @@ class ConversationMessageTask:
         )
 
     def init(self):
+        provider_name = LLMBuilder.get_default_provider(self.app.tenant_id)
+        self.model_dict['provider'] = provider_name
+
         override_model_configs = None
         if self.is_override:
             override_model_configs = {
@@ -79,7 +80,10 @@ class ConversationMessageTask:
             if introduction:
                 prompt_template = OutLinePromptTemplate.from_template(template=PromptBuilder.process_template(introduction))
                 prompt_inputs = {k: self.inputs[k] for k in prompt_template.input_variables if k in self.inputs}
-                introduction = prompt_template.format(**prompt_inputs)
+                try:
+                    introduction = prompt_template.format(**prompt_inputs)
+                except KeyError:
+                    pass
 
             if self.app_model_config.pre_prompt:
                 pre_prompt = PromptBuilder.process_template(self.app_model_config.pre_prompt)
@@ -170,7 +174,7 @@ class ConversationMessageTask:
         )
 
         if not by_stopped:
-            self._pub_handler.pub_end()
+            self.end()
 
     def update_provider_quota(self):
         llm_provider_service = LLMProviderService(
@@ -267,9 +271,12 @@ class ConversationMessageTask:
         total_price = message_tokens_per_1k * message_unit_price + answer_tokens_per_1k * answer_unit_price
         return total_price.quantize(decimal.Decimal('0.0000001'), rounding=decimal.ROUND_HALF_UP)
 
+    def end(self):
+        self._pub_handler.pub_end()
+
 
 class PubHandler:
-    def __init__(self, user: Union[Account | User], task_id: str,
+    def __init__(self, user: Union[Account | EndUser], task_id: str,
                  message: Message, conversation: Conversation,
                  chain_pub: bool = False, agent_thought_pub: bool = False):
         self._channel = PubHandler.generate_channel_name(user, task_id)
@@ -282,12 +289,15 @@ class PubHandler:
         self._agent_thought_pub = agent_thought_pub
 
     @classmethod
-    def generate_channel_name(cls, user: Union[Account | User], task_id: str):
+    def generate_channel_name(cls, user: Union[Account | EndUser], task_id: str):
+        if not user:
+            raise ValueError("user is required")
+
         user_str = 'account-' + user.id if isinstance(user, Account) else 'end-user-' + user.id
         return "generate_result:{}-{}".format(user_str, task_id)
 
     @classmethod
-    def generate_stopped_cache_key(cls, user: Union[Account | User], task_id: str):
+    def generate_stopped_cache_key(cls, user: Union[Account | EndUser], task_id: str):
         user_str = 'account-' + user.id if isinstance(user, Account) else 'end-user-' + user.id
         return "generate_result_stopped:{}-{}".format(user_str, task_id)
 
@@ -366,7 +376,7 @@ class PubHandler:
         redis_client.publish(self._channel, json.dumps(content))
 
     @classmethod
-    def pub_error(cls, user: Union[Account | User], task_id: str, e):
+    def pub_error(cls, user: Union[Account | EndUser], task_id: str, e):
         content = {
             'error': type(e).__name__,
             'description': e.description if getattr(e, 'description', None) is not None else str(e)
@@ -379,7 +389,7 @@ class PubHandler:
         return redis_client.get(self._stopped_cache_key) is not None
 
     @classmethod
-    def stop(cls, user: Union[Account | User], task_id: str):
+    def stop(cls, user: Union[Account | EndUser], task_id: str):
         stopped_cache_key = cls.generate_stopped_cache_key(user, task_id)
         redis_client.setex(stopped_cache_key, 600, 1)
 

api/core/data_source/notion.py

@@ -0,0 +1,369 @@
+"""Notion reader."""
+import json
+import logging
+import os
+from datetime import datetime
+from typing import Any, Dict, List, Optional
+
+import requests  # type: ignore
+
+from llama_index.readers.base import BaseReader
+from llama_index.readers.schema.base import Document
+
+INTEGRATION_TOKEN_NAME = "NOTION_INTEGRATION_TOKEN"
+BLOCK_CHILD_URL_TMPL = "https://api.notion.com/v1/blocks/{block_id}/children"
+DATABASE_URL_TMPL = "https://api.notion.com/v1/databases/{database_id}/query"
+SEARCH_URL = "https://api.notion.com/v1/search"
+RETRIEVE_PAGE_URL_TMPL = "https://api.notion.com/v1/pages/{page_id}"
+RETRIEVE_DATABASE_URL_TMPL = "https://api.notion.com/v1/databases/{database_id}"
+HEADING_TYPE = ['heading_1', 'heading_2', 'heading_3']
+logger = logging.getLogger(__name__)
+
+
+# TODO: Notion DB reader coming soon!
+class NotionPageReader(BaseReader):
+    """Notion Page reader.
+
+    Reads a set of Notion pages.
+
+    Args:
+        integration_token (str): Notion integration token.
+
+    """
+
+    def __init__(self, integration_token: Optional[str] = None) -> None:
+        """Initialize with parameters."""
+        if integration_token is None:
+            integration_token = os.getenv(INTEGRATION_TOKEN_NAME)
+            if integration_token is None:
+                raise ValueError(
+                    "Must specify `integration_token` or set environment "
+                    "variable `NOTION_INTEGRATION_TOKEN`."
+                )
+        self.token = integration_token
+        self.headers = {
+            "Authorization": "Bearer " + self.token,
+            "Content-Type": "application/json",
+            "Notion-Version": "2022-06-28",
+        }
+
+    def _read_block(self, block_id: str, num_tabs: int = 0) -> str:
+        """Read a block."""
+        done = False
+        result_lines_arr = []
+        cur_block_id = block_id
+        while not done:
+            block_url = BLOCK_CHILD_URL_TMPL.format(block_id=cur_block_id)
+            query_dict: Dict[str, Any] = {}
+
+            res = requests.request(
+                "GET", block_url, headers=self.headers, json=query_dict
+            )
+            data = res.json()
+            if 'results' not in data or data["results"] is None:
+                done = True
+                break
+            heading = ''
+            for result in data["results"]:
+                result_type = result["type"]
+                result_obj = result[result_type]
+                cur_result_text_arr = []
+                if result_type == 'table':
+                    result_block_id = result["id"]
+                    text = self._read_table_rows(result_block_id)
+                    result_lines_arr.append(text)
+                else:
+                    if "rich_text" in result_obj:
+                        for rich_text in result_obj["rich_text"]:
+                            # skip if doesn't have text object
+                            if "text" in rich_text:
+                                text = rich_text["text"]["content"]
+                                prefix = "\t" * num_tabs
+                                cur_result_text_arr.append(prefix + text)
+                                if result_type in HEADING_TYPE:
+                                    heading = text
+                    result_block_id = result["id"]
+                    has_children = result["has_children"]
+                    block_type = result["type"]
+                    if has_children and block_type != 'child_page':
+                        children_text = self._read_block(
+                            result_block_id, num_tabs=num_tabs + 1
+                        )
+                        cur_result_text_arr.append(children_text)
+
+                    cur_result_text = "\n".join(cur_result_text_arr)
+                    if result_type in HEADING_TYPE:
+                        result_lines_arr.append(cur_result_text)
+                    else:
+                        result_lines_arr.append(f'{heading}\n{cur_result_text}')
+
+            if data["next_cursor"] is None:
+                done = True
+                break
+            else:
+                cur_block_id = data["next_cursor"]
+
+        result_lines = "\n".join(result_lines_arr)
+        return result_lines
+
+    def _read_table_rows(self, block_id: str) -> str:
+        """Read table rows."""
+        done = False
+        result_lines_arr = []
+        cur_block_id = block_id
+        while not done:
+            block_url = BLOCK_CHILD_URL_TMPL.format(block_id=cur_block_id)
+            query_dict: Dict[str, Any] = {}
+
+            res = requests.request(
+                "GET", block_url, headers=self.headers, json=query_dict
+            )
+            data = res.json()
+            # get table headers text
+            table_header_cell_texts = []
+            tabel_header_cells = data["results"][0]['table_row']['cells']
+            for tabel_header_cell in tabel_header_cells:
+                if tabel_header_cell:
+                    for table_header_cell_text in tabel_header_cell:
+                        text = table_header_cell_text["text"]["content"]
+                        table_header_cell_texts.append(text)
+            # get table columns text and format
+            results = data["results"]
+            for i in range(len(results)-1):
+                column_texts = []
+                tabel_column_cells = data["results"][i+1]['table_row']['cells']
+                for j in range(len(tabel_column_cells)):
+                    if tabel_column_cells[j]:
+                        for table_column_cell_text in tabel_column_cells[j]:
+                            column_text = table_column_cell_text["text"]["content"]
+                            column_texts.append(f'{table_header_cell_texts[j]}:{column_text}')
+
+                cur_result_text = "\n".join(column_texts)
+                result_lines_arr.append(cur_result_text)
+
+            if data["next_cursor"] is None:
+                done = True
+                break
+            else:
+                cur_block_id = data["next_cursor"]
+
+        result_lines = "\n".join(result_lines_arr)
+        return result_lines
+    def _read_parent_blocks(self, block_id: str, num_tabs: int = 0) -> List[str]:
+        """Read a block."""
+        done = False
+        result_lines_arr = []
+        cur_block_id = block_id
+        while not done:
+            block_url = BLOCK_CHILD_URL_TMPL.format(block_id=cur_block_id)
+            query_dict: Dict[str, Any] = {}
+
+            res = requests.request(
+                "GET", block_url, headers=self.headers, json=query_dict
+            )
+            data = res.json()
+            # current block's heading
+            heading = ''
+            for result in data["results"]:
+                result_type = result["type"]
+                result_obj = result[result_type]
+                cur_result_text_arr = []
+                if result_type == 'table':
+                    result_block_id = result["id"]
+                    text = self._read_table_rows(result_block_id)
+                    text += "\n\n"
+                    result_lines_arr.append(text)
+                else:
+                    if "rich_text" in result_obj:
+                        for rich_text in result_obj["rich_text"]:
+                            # skip if doesn't have text object
+                            if "text" in rich_text:
+                                text = rich_text["text"]["content"]
+                                cur_result_text_arr.append(text)
+                                if result_type in HEADING_TYPE:
+                                    heading = text
+
+                    result_block_id = result["id"]
+                    has_children = result["has_children"]
+                    block_type = result["type"]
+                    if has_children and block_type != 'child_page':
+                        children_text = self._read_block(
+                            result_block_id, num_tabs=num_tabs + 1
+                        )
+                        cur_result_text_arr.append(children_text)
+
+                    cur_result_text = "\n".join(cur_result_text_arr)
+                    cur_result_text += "\n\n"
+                    if result_type in HEADING_TYPE:
+                        result_lines_arr.append(cur_result_text)
+                    else:
+                        result_lines_arr.append(f'{heading}\n{cur_result_text}')
+
+            if data["next_cursor"] is None:
+                done = True
+                break
+            else:
+                cur_block_id = data["next_cursor"]
+        return result_lines_arr
+
+    def read_page(self, page_id: str) -> str:
+        """Read a page."""
+        return self._read_block(page_id)
+
+    def read_page_as_documents(self, page_id: str) -> List[str]:
+        """Read a page as documents."""
+        return self._read_parent_blocks(page_id)
+
+    def query_database_data(
+            self, database_id: str, query_dict: Dict[str, Any] = {}
+    ) -> str:
+        """Get all the pages from a Notion database."""
+        res = requests.post\
+                (
+            DATABASE_URL_TMPL.format(database_id=database_id),
+            headers=self.headers,
+            json=query_dict,
+        )
+        data = res.json()
+        database_content_list = []
+        if 'results' not in data or data["results"] is None:
+            return ""
+        for result in data["results"]:
+            properties = result['properties']
+            data = {}
+            for property_name, property_value in properties.items():
+                type = property_value['type']
+                if type == 'multi_select':
+                    value = []
+                    multi_select_list = property_value[type]
+                    for multi_select in multi_select_list:
+                        value.append(multi_select['name'])
+                elif type == 'rich_text' or type == 'title':
+                    if len(property_value[type]) > 0:
+                        value = property_value[type][0]['plain_text']
+                    else:
+                        value = ''
+                elif type == 'select' or type == 'status':
+                    if property_value[type]:
+                        value = property_value[type]['name']
+                    else:
+                        value = ''
+                else:
+                    value = property_value[type]
+                data[property_name] = value
+            database_content_list.append(json.dumps(data))
+
+        return "\n\n".join(database_content_list)
+
+    def query_database(
+            self, database_id: str, query_dict: Dict[str, Any] = {}
+    ) -> List[str]:
+        """Get all the pages from a Notion database."""
+        res = requests.post\
+                (
+            DATABASE_URL_TMPL.format(database_id=database_id),
+            headers=self.headers,
+            json=query_dict,
+        )
+        data = res.json()
+        page_ids = []
+        for result in data["results"]:
+            page_id = result["id"]
+            page_ids.append(page_id)
+
+        return page_ids
+
+    def search(self, query: str) -> List[str]:
+        """Search Notion page given a text query."""
+        done = False
+        next_cursor: Optional[str] = None
+        page_ids = []
+        while not done:
+            query_dict = {
+                "query": query,
+            }
+            if next_cursor is not None:
+                query_dict["start_cursor"] = next_cursor
+            res = requests.post(SEARCH_URL, headers=self.headers, json=query_dict)
+            data = res.json()
+            for result in data["results"]:
+                page_id = result["id"]
+                page_ids.append(page_id)
+
+            if data["next_cursor"] is None:
+                done = True
+                break
+            else:
+                next_cursor = data["next_cursor"]
+        return page_ids
+
+    def load_data(
+            self, page_ids: List[str] = [], database_id: Optional[str] = None
+    ) -> List[Document]:
+        """Load data from the input directory.
+
+        Args:
+            page_ids (List[str]): List of page ids to load.
+
+        Returns:
+            List[Document]: List of documents.
+
+        """
+        if not page_ids and not database_id:
+            raise ValueError("Must specify either `page_ids` or `database_id`.")
+        docs = []
+        if database_id is not None:
+            # get all the pages in the database
+            page_ids = self.query_database(database_id)
+            for page_id in page_ids:
+                page_text = self.read_page(page_id)
+                docs.append(Document(page_text))
+        else:
+            for page_id in page_ids:
+                page_text = self.read_page(page_id)
+                docs.append(Document(page_text))
+
+        return docs
+
+    def load_data_as_documents(
+            self, page_ids: List[str] = [], database_id: Optional[str] = None
+    ) -> List[Document]:
+        if not page_ids and not database_id:
+            raise ValueError("Must specify either `page_ids` or `database_id`.")
+        docs = []
+        if database_id is not None:
+            # get all the pages in the database
+            page_text = self.query_database_data(database_id)
+            docs.append(Document(page_text))
+        else:
+            for page_id in page_ids:
+                page_text_list = self.read_page_as_documents(page_id)
+                for page_text in page_text_list:
+                    docs.append(Document(page_text))
+
+        return docs
+
+    def get_page_last_edited_time(self, page_id: str) -> str:
+        retrieve_page_url = RETRIEVE_PAGE_URL_TMPL.format(page_id=page_id)
+        query_dict: Dict[str, Any] = {}
+
+        res = requests.request(
+            "GET", retrieve_page_url, headers=self.headers, json=query_dict
+        )
+        data = res.json()
+        return data["last_edited_time"]
+
+    def get_database_last_edited_time(self, database_id: str) -> str:
+        retrieve_page_url = RETRIEVE_DATABASE_URL_TMPL.format(database_id=database_id)
+        query_dict: Dict[str, Any] = {}
+
+        res = requests.request(
+            "GET", retrieve_page_url, headers=self.headers, json=query_dict
+        )
+        data = res.json()
+        return data["last_edited_time"]
+
+
+if __name__ == "__main__":
+    reader = NotionPageReader()
+    logger.info(reader.search("What I"))

api/core/embedding/openai_embedding.py

@@ -11,9 +11,10 @@ from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_except
 
 @retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6))
 def get_embedding(
-    text: str,
-    engine: Optional[str] = None,
-    openai_api_key: Optional[str] = None,
+        text: str,
+        engine: Optional[str] = None,
+        api_key: Optional[str] = None,
+        **kwargs
 ) -> List[float]:
     """Get embedding.
 
@@ -25,11 +26,12 @@ def get_embedding(
 
     """
     text = text.replace("\n", " ")
-    return openai.Embedding.create(input=[text], engine=engine, api_key=openai_api_key)["data"][0]["embedding"]
+    return openai.Embedding.create(input=[text], engine=engine, api_key=api_key, **kwargs)["data"][0]["embedding"]
 
 
 @retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6))
-async def aget_embedding(text: str, engine: Optional[str] = None, openai_api_key: Optional[str] = None) -> List[float]:
+async def aget_embedding(text: str, engine: Optional[str] = None, api_key: Optional[str] = None, **kwargs) -> List[
+    float]:
     """Asynchronously get embedding.
 
     NOTE: Copied from OpenAI's embedding utils:
@@ -42,16 +44,17 @@ async def aget_embedding(text: str, engine: Optional[str] = None, openai_api_key
     # replace newlines, which can negatively affect performance.
     text = text.replace("\n", " ")
 
-    return (await openai.Embedding.acreate(input=[text], engine=engine, api_key=openai_api_key))["data"][0][
+    return (await openai.Embedding.acreate(input=[text], engine=engine, api_key=api_key, **kwargs))["data"][0][
         "embedding"
     ]
 
 
 @retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6))
 def get_embeddings(
-    list_of_text: List[str],
-    engine: Optional[str] = None,
-    openai_api_key: Optional[str] = None
+        list_of_text: List[str],
+        engine: Optional[str] = None,
+        api_key: Optional[str] = None,
+        **kwargs
 ) -> List[List[float]]:
     """Get embeddings.
 
@@ -67,14 +70,14 @@ def get_embeddings(
     # replace newlines, which can negatively affect performance.
     list_of_text = [text.replace("\n", " ") for text in list_of_text]
 
-    data = openai.Embedding.create(input=list_of_text, engine=engine, api_key=openai_api_key).data
+    data = openai.Embedding.create(input=list_of_text, engine=engine, api_key=api_key, **kwargs).data
     data = sorted(data, key=lambda x: x["index"])  # maintain the same order as input.
     return [d["embedding"] for d in data]
 
 
 @retry(reraise=True, wait=wait_random_exponential(min=1, max=20), stop=stop_after_attempt(6))
 async def aget_embeddings(
-    list_of_text: List[str], engine: Optional[str] = None, openai_api_key: Optional[str] = None
+        list_of_text: List[str], engine: Optional[str] = None, api_key: Optional[str] = None, **kwargs
 ) -> List[List[float]]:
     """Asynchronously get embeddings.
 
@@ -90,7 +93,7 @@ async def aget_embeddings(
     # replace newlines, which can negatively affect performance.
     list_of_text = [text.replace("\n", " ") for text in list_of_text]
 
-    data = (await openai.Embedding.acreate(input=list_of_text, engine=engine, api_key=openai_api_key)).data
+    data = (await openai.Embedding.acreate(input=list_of_text, engine=engine, api_key=api_key, **kwargs)).data
     data = sorted(data, key=lambda x: x["index"])  # maintain the same order as input.
     return [d["embedding"] for d in data]
 
@@ -98,19 +101,30 @@ async def aget_embeddings(
 class OpenAIEmbedding(BaseEmbedding):
 
     def __init__(
-        self,
-        mode: str = OpenAIEmbeddingMode.TEXT_SEARCH_MODE,
-        model: str = OpenAIEmbeddingModelType.TEXT_EMBED_ADA_002,
-        deployment_name: Optional[str] = None,
-        openai_api_key: Optional[str] = None,
-        **kwargs: Any,
+            self,
+            mode: str = OpenAIEmbeddingMode.TEXT_SEARCH_MODE,
+            model: str = OpenAIEmbeddingModelType.TEXT_EMBED_ADA_002,
+            deployment_name: Optional[str] = None,
+            openai_api_key: Optional[str] = None,
+            **kwargs: Any,
     ) -> None:
         """Init params."""
-        super().__init__(**kwargs)
+        new_kwargs = {}
+
+        if 'embed_batch_size' in kwargs:
+            new_kwargs['embed_batch_size'] = kwargs['embed_batch_size']
+
+        if 'tokenizer' in kwargs:
+            new_kwargs['tokenizer'] = kwargs['tokenizer']
+
+        super().__init__(**new_kwargs)
         self.mode = OpenAIEmbeddingMode(mode)
         self.model = OpenAIEmbeddingModelType(model)
         self.deployment_name = deployment_name
         self.openai_api_key = openai_api_key
+        self.openai_api_type = kwargs.get('openai_api_type')
+        self.openai_api_version = kwargs.get('openai_api_version')
+        self.openai_api_base = kwargs.get('openai_api_base')
 
     @handle_llm_exceptions
     def _get_query_embedding(self, query: str) -> List[float]:
@@ -122,7 +136,9 @@ class OpenAIEmbedding(BaseEmbedding):
             if key not in _QUERY_MODE_MODEL_DICT:
                 raise ValueError(f"Invalid mode, model combination: {key}")
             engine = _QUERY_MODE_MODEL_DICT[key]
-        return get_embedding(query, engine=engine, openai_api_key=self.openai_api_key)
+        return get_embedding(query, engine=engine, api_key=self.openai_api_key,
+                             api_type=self.openai_api_type, api_version=self.openai_api_version,
+                             api_base=self.openai_api_base)
 
     def _get_text_embedding(self, text: str) -> List[float]:
         """Get text embedding."""
@@ -133,7 +149,9 @@ class OpenAIEmbedding(BaseEmbedding):
             if key not in _TEXT_MODE_MODEL_DICT:
                 raise ValueError(f"Invalid mode, model combination: {key}")
             engine = _TEXT_MODE_MODEL_DICT[key]
-        return get_embedding(text, engine=engine, openai_api_key=self.openai_api_key)
+        return get_embedding(text, engine=engine, api_key=self.openai_api_key,
+                             api_type=self.openai_api_type, api_version=self.openai_api_version,
+                             api_base=self.openai_api_base)
 
     async def _aget_text_embedding(self, text: str) -> List[float]:
         """Asynchronously get text embedding."""
@@ -144,7 +162,9 @@ class OpenAIEmbedding(BaseEmbedding):
             if key not in _TEXT_MODE_MODEL_DICT:
                 raise ValueError(f"Invalid mode, model combination: {key}")
             engine = _TEXT_MODE_MODEL_DICT[key]
-        return await aget_embedding(text, engine=engine, openai_api_key=self.openai_api_key)
+        return await aget_embedding(text, engine=engine, api_key=self.openai_api_key,
+                                    api_type=self.openai_api_type, api_version=self.openai_api_version,
+                                    api_base=self.openai_api_base)
 
     def _get_text_embeddings(self, texts: List[str]) -> List[List[float]]:
         """Get text embeddings.
@@ -153,6 +173,13 @@ class OpenAIEmbedding(BaseEmbedding):
         Can be overriden for batch queries.
 
         """
+        if self.openai_api_type and self.openai_api_type == 'azure':
+            embeddings = []
+            for text in texts:
+                embeddings.append(self._get_text_embedding(text))
+
+            return embeddings
+
         if self.deployment_name is not None:
             engine = self.deployment_name
         else:
@@ -160,11 +187,20 @@ class OpenAIEmbedding(BaseEmbedding):
             if key not in _TEXT_MODE_MODEL_DICT:
                 raise ValueError(f"Invalid mode, model combination: {key}")
             engine = _TEXT_MODE_MODEL_DICT[key]
-        embeddings = get_embeddings(texts, engine=engine, openai_api_key=self.openai_api_key)
+        embeddings = get_embeddings(texts, engine=engine, api_key=self.openai_api_key,
+                                    api_type=self.openai_api_type, api_version=self.openai_api_version,
+                                    api_base=self.openai_api_base)
         return embeddings
 
     async def _aget_text_embeddings(self, texts: List[str]) -> List[List[float]]:
         """Asynchronously get text embeddings."""
+        if self.openai_api_type and self.openai_api_type == 'azure':
+            embeddings = []
+            for text in texts:
+                embeddings.append(await self._aget_text_embedding(text))
+
+            return embeddings
+
         if self.deployment_name is not None:
             engine = self.deployment_name
         else:
@@ -172,5 +208,7 @@ class OpenAIEmbedding(BaseEmbedding):
             if key not in _TEXT_MODE_MODEL_DICT:
                 raise ValueError(f"Invalid mode, model combination: {key}")
             engine = _TEXT_MODE_MODEL_DICT[key]
-        embeddings = await aget_embeddings(texts, engine=engine, openai_api_key=self.openai_api_key)
+        embeddings = await aget_embeddings(texts, engine=engine, api_key=self.openai_api_key,
+                                           api_type=self.openai_api_type, api_version=self.openai_api_version,
+                                           api_base=self.openai_api_base)
         return embeddings

api/core/generator/llm_generator.py

@@ -1,12 +1,13 @@
 import logging
 
 from langchain.chat_models.base import BaseChatModel
-from langchain.schema import HumanMessage
+from langchain.schema import HumanMessage, OutputParserException
 
 from core.constant import llm_constant
 from core.llm.llm_builder import LLMBuilder
 from core.llm.streamable_open_ai import StreamableOpenAI
 from core.llm.token_calculator import TokenCalculator
+from core.prompt.output_parser.rule_config_generator import RuleConfigGeneratorOutputParser
 
 from core.prompt.output_parser.suggested_questions_after_answer import SuggestedQuestionsAfterAnswerOutputParser
 from core.prompt.prompt_template import OutLinePromptTemplate
@@ -118,3 +119,48 @@ class LLMGenerator:
             questions = []
 
         return questions
+
+    @classmethod
+    def generate_rule_config(cls, tenant_id: str, audiences: str, hoping_to_solve: str) -> dict:
+        output_parser = RuleConfigGeneratorOutputParser()
+
+        prompt = OutLinePromptTemplate(
+            template=output_parser.get_format_instructions(),
+            input_variables=["audiences", "hoping_to_solve"],
+            partial_variables={
+                "variable": '{variable}',
+                "lanA": '{lanA}',
+                "lanB": '{lanB}',
+                "topic": '{topic}'
+            },
+            validate_template=False
+        )
+
+        _input = prompt.format_prompt(audiences=audiences, hoping_to_solve=hoping_to_solve)
+
+        llm: StreamableOpenAI = LLMBuilder.to_llm(
+            tenant_id=tenant_id,
+            model_name=generate_base_model,
+            temperature=0,
+            max_tokens=512
+        )
+
+        if isinstance(llm, BaseChatModel):
+            query = [HumanMessage(content=_input.to_string())]
+        else:
+            query = _input.to_string()
+
+        try:
+            output = llm(query)
+            rule_config = output_parser.parse(output)
+        except OutputParserException:
+            raise ValueError('Please give a valid input for intended audience or hoping to solve problems.')
+        except Exception:
+            logging.exception("Error generating prompt")
+            rule_config = {
+                "prompt": "",
+                "variables": [],
+                "opening_statement": ""
+            }
+
+        return rule_config

api/core/index/index_builder.py

@@ -33,8 +33,11 @@ class IndexBuilder:
             max_chunk_overlap=20
         )
 
+        provider = LLMBuilder.get_default_provider(tenant_id)
+
         model_credentials = LLMBuilder.get_model_credentials(
             tenant_id=tenant_id,
+            model_provider=provider,
             model_name='text-embedding-ada-002'
         )
 
@@ -43,3 +46,15 @@ class IndexBuilder:
             prompt_helper=prompt_helper,
             embed_model=OpenAIEmbedding(**model_credentials),
         )
+
+    @classmethod
+    def get_fake_llm_service_context(cls, tenant_id: str) -> ServiceContext:
+        llm = LLMBuilder.to_llm(
+            tenant_id=tenant_id,
+            model_name='fake'
+        )
+
+        return ServiceContext.from_defaults(
+            llm_predictor=LLMPredictor(llm=llm),
+            embed_model=OpenAIEmbedding()
+        )

api/core/index/readers/markdown_parser.py

@@ -0,0 +1,111 @@
+"""Markdown parser.
+
+Contains parser for md files.
+
+"""
+import re
+from pathlib import Path
+from typing import Any, Dict, List, Optional, Tuple, Union, cast
+
+from llama_index.readers.file.base_parser import BaseParser
+
+
+class MarkdownParser(BaseParser):
+    """Markdown parser.
+
+    Extract text from markdown files.
+    Returns dictionary with keys as headers and values as the text between headers.
+
+    """
+
+    def __init__(
+        self,
+        *args: Any,
+        remove_hyperlinks: bool = True,
+        remove_images: bool = True,
+        **kwargs: Any,
+    ) -> None:
+        """Init params."""
+        super().__init__(*args, **kwargs)
+        self._remove_hyperlinks = remove_hyperlinks
+        self._remove_images = remove_images
+
+    def markdown_to_tups(self, markdown_text: str) -> List[Tuple[Optional[str], str]]:
+        """Convert a markdown file to a dictionary.
+
+        The keys are the headers and the values are the text under each header.
+
+        """
+        markdown_tups: List[Tuple[Optional[str], str]] = []
+        lines = markdown_text.split("\n")
+
+        current_header = None
+        current_text = ""
+
+        for line in lines:
+            header_match = re.match(r"^#+\s", line)
+            if header_match:
+                if current_header is not None:
+                    markdown_tups.append((current_header, current_text))
+
+                current_header = line
+                current_text = ""
+            else:
+                current_text += line + "\n"
+        markdown_tups.append((current_header, current_text))
+
+        if current_header is not None:
+            # pass linting, assert keys are defined
+            markdown_tups = [
+                (re.sub(r"#", "", cast(str, key)).strip(), re.sub(r"<.*?>", "", value))
+                for key, value in markdown_tups
+            ]
+        else:
+            markdown_tups = [
+                (key, re.sub("\n", "", value)) for key, value in markdown_tups
+            ]
+
+        return markdown_tups
+
+    def remove_images(self, content: str) -> str:
+        """Get a dictionary of a markdown file from its path."""
+        pattern = r"!{1}\[\[(.*)\]\]"
+        content = re.sub(pattern, "", content)
+        return content
+
+    def remove_hyperlinks(self, content: str) -> str:
+        """Get a dictionary of a markdown file from its path."""
+        pattern = r"\[(.*?)\]\((.*?)\)"
+        content = re.sub(pattern, r"\1", content)
+        return content
+
+    def _init_parser(self) -> Dict:
+        """Initialize the parser with the config."""
+        return {}
+
+    def parse_tups(
+        self, filepath: Path, errors: str = "ignore"
+    ) -> List[Tuple[Optional[str], str]]:
+        """Parse file into tuples."""
+        with open(filepath, "r", encoding="utf-8") as f:
+            content = f.read()
+        if self._remove_hyperlinks:
+            content = self.remove_hyperlinks(content)
+        if self._remove_images:
+            content = self.remove_images(content)
+        markdown_tups = self.markdown_to_tups(content)
+        return markdown_tups
+
+    def parse_file(
+        self, filepath: Path, errors: str = "ignore"
+    ) -> Union[str, List[str]]:
+        """Parse file into string."""
+        tups = self.parse_tups(filepath, errors=errors)
+        results = []
+        # TODO: don't include headers right now
+        for header, value in tups:
+            if header is None:
+                results.append(value)
+            else:
+                results.append(f"\n\n{header}\n{value}")
+        return results

api/core/index/readers/xlsx_parser.py

@@ -0,0 +1,31 @@
+from pathlib import Path
+import json
+from typing import Dict
+from openpyxl import load_workbook
+
+from llama_index.readers.file.base_parser import BaseParser
+from flask import current_app
+
+
+class XLSXParser(BaseParser):
+    """XLSX parser."""
+
+    def _init_parser(self) -> Dict:
+        """Init parser"""
+        return {}
+
+    def parse_file(self, file: Path, errors: str = "ignore") -> str:
+        data = []
+        keys = []
+        with open(file, "r") as fp:
+            wb = load_workbook(filename=file, read_only=True)
+            # loop over all sheets
+            for sheet in wb:
+                for row in sheet.iter_rows(values_only=True):
+                    if all(v is None for v in row):
+                        continue
+                    if keys == []:
+                        keys = list(map(str, row))
+                    else:
+                        data.append(json.dumps(dict(zip(keys, list(map(str, row)))), ensure_ascii=False))
+        return '\n\n'.join(data)

api/core/index/spiltter/fixed_text_splitter.py

@@ -0,0 +1,68 @@
+"""Functionality for splitting text."""
+from __future__ import annotations
+
+from typing import (
+    Any,
+    List,
+    Optional,
+)
+
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+
+
+class FixedRecursiveCharacterTextSplitter(RecursiveCharacterTextSplitter):
+    def __init__(self, fixed_separator: str = "\n\n", separators: Optional[List[str]] = None, **kwargs: Any):
+        """Create a new TextSplitter."""
+        super().__init__(**kwargs)
+        self._fixed_separator = fixed_separator
+        self._separators = separators or ["\n\n", "\n", " ", ""]
+
+    def split_text(self, text: str) -> List[str]:
+        """Split incoming text and return chunks."""
+        if self._fixed_separator:
+            chunks = text.split(self._fixed_separator)
+        else:
+            chunks = list(text)
+
+        final_chunks = []
+        for chunk in chunks:
+            if self._length_function(chunk) > self._chunk_size:
+                final_chunks.extend(self.recursive_split_text(chunk))
+            else:
+                final_chunks.append(chunk)
+
+        return final_chunks
+
+    def recursive_split_text(self, text: str) -> List[str]:
+        """Split incoming text and return chunks."""
+        final_chunks = []
+        # Get appropriate separator to use
+        separator = self._separators[-1]
+        for _s in self._separators:
+            if _s == "":
+                separator = _s
+                break
+            if _s in text:
+                separator = _s
+                break
+        # Now that we have the separator, split the text
+        if separator:
+            splits = text.split(separator)
+        else:
+            splits = list(text)
+        # Now go merging things, recursively splitting longer texts.
+        _good_splits = []
+        for s in splits:
+            if self._length_function(s) < self._chunk_size:
+                _good_splits.append(s)
+            else:
+                if _good_splits:
+                    merged_text = self._merge_splits(_good_splits, separator)
+                    final_chunks.extend(merged_text)
+                    _good_splits = []
+                other_info = self.recursive_split_text(s)
+                final_chunks.extend(other_info)
+        if _good_splits:
+            merged_text = self._merge_splits(_good_splits, separator)
+            final_chunks.extend(merged_text)
+        return final_chunks

api/core/index/vector_index.py

@@ -83,7 +83,7 @@ class VectorIndex:
         if not self._dataset.index_struct_dict:
             return
 
-        service_context = IndexBuilder.get_default_service_context(tenant_id=self._dataset.tenant_id)
+        service_context = IndexBuilder.get_fake_llm_service_context(tenant_id=self._dataset.tenant_id)
 
         index = vector_store.get_index(
             service_context=service_context,
@@ -101,7 +101,7 @@ class VectorIndex:
         if not self._dataset.index_struct_dict:
             return
 
-        service_context = IndexBuilder.get_default_service_context(tenant_id=self._dataset.tenant_id)
+        service_context = IndexBuilder.get_fake_llm_service_context(tenant_id=self._dataset.tenant_id)
 
         index = vector_store.get_index(
             service_context=service_context,

api/core/indexing_runner.py

@@ -5,6 +5,8 @@ import tempfile
 import time
 from pathlib import Path
 from typing import Optional, List
+
+from flask_login import current_user
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 
 from llama_index import SimpleDirectoryReader
@@ -14,10 +16,14 @@ from llama_index.node_parser import SimpleNodeParser, NodeParser
 from llama_index.readers.file.base import DEFAULT_FILE_EXTRACTOR
 from llama_index.readers.file.markdown_parser import MarkdownParser
 
+from core.data_source.notion import NotionPageReader
+from core.index.readers.xlsx_parser import XLSXParser
 from core.docstore.dataset_docstore import DatesetDocumentStore
 from core.index.keyword_table_index import KeywordTableIndex
 from core.index.readers.html_parser import HTMLParser
+from core.index.readers.markdown_parser import MarkdownParser
 from core.index.readers.pdf_parser import PDFParser
+from core.index.spiltter.fixed_text_splitter import FixedRecursiveCharacterTextSplitter
 from core.index.vector_index import VectorIndex
 from core.llm.token_calculator import TokenCalculator
 from extensions.ext_database import db
@@ -25,6 +31,7 @@ from extensions.ext_redis import redis_client
 from extensions.ext_storage import storage
 from models.dataset import Document, Dataset, DocumentSegment, DatasetProcessRule
 from models.model import UploadFile
+from models.source import DataSourceBinding
 
 
 class IndexingRunner:
@@ -33,42 +40,43 @@ class IndexingRunner:
         self.storage = storage
         self.embedding_model_name = embedding_model_name
 
-    def run(self, document: Document):
+    def run(self, documents: List[Document]):
         """Run the indexing process."""
-        # get dataset
-        dataset = Dataset.query.filter_by(
-            id=document.dataset_id
-        ).first()
+        for document in documents:
+            # get dataset
+            dataset = Dataset.query.filter_by(
+                id=document.dataset_id
+            ).first()
 
-        if not dataset:
-            raise ValueError("no dataset found")
+            if not dataset:
+                raise ValueError("no dataset found")
 
-        # load file
-        text_docs = self._load_data(document)
+            # load file
+            text_docs = self._load_data(document)
 
-        # get the process rule
-        processing_rule = db.session.query(DatasetProcessRule). \
-            filter(DatasetProcessRule.id == document.dataset_process_rule_id). \
-            first()
+            # get the process rule
+            processing_rule = db.session.query(DatasetProcessRule). \
+                filter(DatasetProcessRule.id == document.dataset_process_rule_id). \
+                first()
 
-        # get node parser for splitting
-        node_parser = self._get_node_parser(processing_rule)
+            # get node parser for splitting
+            node_parser = self._get_node_parser(processing_rule)
 
-        # split to nodes
-        nodes = self._step_split(
-            text_docs=text_docs,
-            node_parser=node_parser,
-            dataset=dataset,
-            document=document,
-            processing_rule=processing_rule
-        )
+            # split to nodes
+            nodes = self._step_split(
+                text_docs=text_docs,
+                node_parser=node_parser,
+                dataset=dataset,
+                document=document,
+                processing_rule=processing_rule
+            )
 
-        # build index
-        self._build_index(
-            dataset=dataset,
-            document=document,
-            nodes=nodes
-        )
+            # build index
+            self._build_index(
+                dataset=dataset,
+                document=document,
+                nodes=nodes
+            )
 
     def run_in_splitting_status(self, document: Document):
         """Run the indexing process when the index_status is splitting."""
@@ -162,38 +170,98 @@ class IndexingRunner:
             nodes=nodes
         )
 
-    def indexing_estimate(self, file_detail: UploadFile, tmp_processing_rule: dict) -> dict:
+    def file_indexing_estimate(self, file_details: List[UploadFile], tmp_processing_rule: dict) -> dict:
         """
         Estimate the indexing for the document.
         """
-        # load data from file
-        text_docs = self._load_data_from_file(file_detail)
-
-        processing_rule = DatasetProcessRule(
-            mode=tmp_processing_rule["mode"],
-            rules=json.dumps(tmp_processing_rule["rules"])
-        )
-
-        # get node parser for splitting
-        node_parser = self._get_node_parser(processing_rule)
-
-        # split to nodes
-        nodes = self._split_to_nodes(
-            text_docs=text_docs,
-            node_parser=node_parser,
-            processing_rule=processing_rule
-        )
-
         tokens = 0
         preview_texts = []
-        for node in nodes:
-            if len(preview_texts) < 5:
-                preview_texts.append(node.get_text())
+        total_segments = 0
+        for file_detail in file_details:
+            # load data from file
+            text_docs = self._load_data_from_file(file_detail)
 
-            tokens += TokenCalculator.get_num_tokens(self.embedding_model_name, node.get_text())
+            processing_rule = DatasetProcessRule(
+                mode=tmp_processing_rule["mode"],
+                rules=json.dumps(tmp_processing_rule["rules"])
+            )
+
+            # get node parser for splitting
+            node_parser = self._get_node_parser(processing_rule)
+
+            # split to nodes
+            nodes = self._split_to_nodes(
+                text_docs=text_docs,
+                node_parser=node_parser,
+                processing_rule=processing_rule
+            )
+            total_segments += len(nodes)
+            for node in nodes:
+                if len(preview_texts) < 5:
+                    preview_texts.append(node.get_text())
+
+                tokens += TokenCalculator.get_num_tokens(self.embedding_model_name, node.get_text())
 
         return {
-            "total_segments": len(nodes),
+            "total_segments": total_segments,
+            "tokens": tokens,
+            "total_price": '{:f}'.format(TokenCalculator.get_token_price(self.embedding_model_name, tokens)),
+            "currency": TokenCalculator.get_currency(self.embedding_model_name),
+            "preview": preview_texts
+        }
+
+    def notion_indexing_estimate(self, notion_info_list: list, tmp_processing_rule: dict) -> dict:
+        """
+        Estimate the indexing for the document.
+        """
+        # load data from notion
+        tokens = 0
+        preview_texts = []
+        total_segments = 0
+        for notion_info in notion_info_list:
+            workspace_id = notion_info['workspace_id']
+            data_source_binding = DataSourceBinding.query.filter(
+                db.and_(
+                    DataSourceBinding.tenant_id == current_user.current_tenant_id,
+                    DataSourceBinding.provider == 'notion',
+                    DataSourceBinding.disabled == False,
+                    DataSourceBinding.source_info['workspace_id'] == f'"{workspace_id}"'
+                )
+            ).first()
+            if not data_source_binding:
+                raise ValueError('Data source binding not found.')
+            reader = NotionPageReader(integration_token=data_source_binding.access_token)
+            for page in notion_info['pages']:
+                if page['type'] == 'page':
+                    page_ids = [page['page_id']]
+                    documents = reader.load_data_as_documents(page_ids=page_ids)
+                elif page['type'] == 'database':
+                    documents = reader.load_data_as_documents(database_id=page['page_id'])
+                else:
+                    documents = []
+                processing_rule = DatasetProcessRule(
+                    mode=tmp_processing_rule["mode"],
+                    rules=json.dumps(tmp_processing_rule["rules"])
+                )
+
+                # get node parser for splitting
+                node_parser = self._get_node_parser(processing_rule)
+
+                # split to nodes
+                nodes = self._split_to_nodes(
+                    text_docs=documents,
+                    node_parser=node_parser,
+                    processing_rule=processing_rule
+                )
+                total_segments += len(nodes)
+                for node in nodes:
+                    if len(preview_texts) < 5:
+                        preview_texts.append(node.get_text())
+
+                    tokens += TokenCalculator.get_num_tokens(self.embedding_model_name, node.get_text())
+
+        return {
+            "total_segments": total_segments,
             "tokens": tokens,
             "total_price": '{:f}'.format(TokenCalculator.get_token_price(self.embedding_model_name, tokens)),
             "currency": TokenCalculator.get_currency(self.embedding_model_name),
@@ -202,25 +270,50 @@ class IndexingRunner:
 
     def _load_data(self, document: Document) -> List[Document]:
         # load file
-        if document.data_source_type != "upload_file":
+        if document.data_source_type not in ["upload_file", "notion_import"]:
             return []
 
         data_source_info = document.data_source_info_dict
-        if not data_source_info or 'upload_file_id' not in data_source_info:
-            raise ValueError("no upload file found")
+        text_docs = []
+        if document.data_source_type == 'upload_file':
+            if not data_source_info or 'upload_file_id' not in data_source_info:
+                raise ValueError("no upload file found")
 
-        file_detail = db.session.query(UploadFile). \
-            filter(UploadFile.id == data_source_info['upload_file_id']). \
-            one_or_none()
-
-        text_docs = self._load_data_from_file(file_detail)
+            file_detail = db.session.query(UploadFile). \
+                filter(UploadFile.id == data_source_info['upload_file_id']). \
+                one_or_none()
 
+            text_docs = self._load_data_from_file(file_detail)
+        elif document.data_source_type == 'notion_import':
+            if not data_source_info or 'notion_page_id' not in data_source_info \
+                    or 'notion_workspace_id' not in data_source_info:
+                raise ValueError("no notion page found")
+            workspace_id = data_source_info['notion_workspace_id']
+            page_id = data_source_info['notion_page_id']
+            page_type = data_source_info['type']
+            data_source_binding = DataSourceBinding.query.filter(
+                db.and_(
+                    DataSourceBinding.tenant_id == document.tenant_id,
+                    DataSourceBinding.provider == 'notion',
+                    DataSourceBinding.disabled == False,
+                    DataSourceBinding.source_info['workspace_id'] == f'"{workspace_id}"'
+                )
+            ).first()
+            if not data_source_binding:
+                raise ValueError('Data source binding not found.')
+            if page_type == 'page':
+                # add page last_edited_time to data_source_info
+                self._get_notion_page_last_edited_time(page_id, data_source_binding.access_token, document)
+                text_docs = self._load_page_data_from_notion(page_id, data_source_binding.access_token)
+            elif page_type == 'database':
+                # add page last_edited_time to data_source_info
+                self._get_notion_database_last_edited_time(page_id, data_source_binding.access_token, document)
+                text_docs = self._load_database_data_from_notion(page_id, data_source_binding.access_token)
         # update document status to splitting
         self._update_document_index_status(
             document_id=document.id,
             after_indexing_status="splitting",
             extra_update_params={
-                Document.file_id: file_detail.id,
                 Document.word_count: sum([len(text_doc.text) for text_doc in text_docs]),
                 Document.parsing_completed_at: datetime.datetime.utcnow()
             }
@@ -246,15 +339,52 @@ class IndexingRunner:
 
             file_extractor = DEFAULT_FILE_EXTRACTOR.copy()
             file_extractor[".markdown"] = MarkdownParser()
+            file_extractor[".md"] = MarkdownParser()
             file_extractor[".html"] = HTMLParser()
             file_extractor[".htm"] = HTMLParser()
             file_extractor[".pdf"] = PDFParser({'upload_file': upload_file})
+            file_extractor[".xlsx"] = XLSXParser()
 
             loader = SimpleDirectoryReader(input_files=[filepath], file_extractor=file_extractor)
             text_docs = loader.load_data()
 
             return text_docs
 
+    def _load_page_data_from_notion(self, page_id: str, access_token: str) -> List[Document]:
+        page_ids = [page_id]
+        reader = NotionPageReader(integration_token=access_token)
+        text_docs = reader.load_data_as_documents(page_ids=page_ids)
+        return text_docs
+
+    def _load_database_data_from_notion(self, database_id: str, access_token: str) -> List[Document]:
+        reader = NotionPageReader(integration_token=access_token)
+        text_docs = reader.load_data_as_documents(database_id=database_id)
+        return text_docs
+
+    def _get_notion_page_last_edited_time(self, page_id: str, access_token: str, document: Document):
+        reader = NotionPageReader(integration_token=access_token)
+        last_edited_time = reader.get_page_last_edited_time(page_id)
+        data_source_info = document.data_source_info_dict
+        data_source_info['last_edited_time'] = last_edited_time
+        update_params = {
+            Document.data_source_info: json.dumps(data_source_info)
+        }
+
+        Document.query.filter_by(id=document.id).update(update_params)
+        db.session.commit()
+
+    def _get_notion_database_last_edited_time(self, page_id: str, access_token: str, document: Document):
+        reader = NotionPageReader(integration_token=access_token)
+        last_edited_time = reader.get_database_last_edited_time(page_id)
+        data_source_info = document.data_source_info_dict
+        data_source_info['last_edited_time'] = last_edited_time
+        update_params = {
+            Document.data_source_info: json.dumps(data_source_info)
+        }
+
+        Document.query.filter_by(id=document.id).update(update_params)
+        db.session.commit()
+
     def _get_node_parser(self, processing_rule: DatasetProcessRule) -> NodeParser:
         """
         Get the NodeParser object according to the processing rule.
@@ -267,16 +397,14 @@ class IndexingRunner:
                 raise ValueError("Custom segment length should be between 50 and 1000.")
 
             separator = segmentation["separator"]
-            if not separator:
-                separators = ["\n\n", "。", ".", " ", ""]
-            else:
+            if separator:
                 separator = separator.replace('\\n', '\n')
-                separators = [separator, ""]
 
-            character_splitter = RecursiveCharacterTextSplitter.from_tiktoken_encoder(
+            character_splitter = FixedRecursiveCharacterTextSplitter.from_tiktoken_encoder(
                 chunk_size=segmentation["max_tokens"],
                 chunk_overlap=0,
-                separators=separators
+                fixed_separator=separator,
+                separators=["\n\n", "。", ".", " ", ""]
             )
         else:
             # Automatic segmentation
@@ -306,7 +434,7 @@ class IndexingRunner:
             embedding_model_name=self.embedding_model_name,
             document_id=document.id
         )
-
+        # add document segments
         doc_store.add_documents(nodes)
 
         # update document status to indexing
@@ -344,7 +472,7 @@ class IndexingRunner:
 
             # parse document to nodes
             nodes = node_parser.get_nodes_from_documents([text_doc])
-
+            nodes = [node for node in nodes if node.text is not None and node.text.strip()]
             all_nodes.extend(nodes)
 
         return all_nodes

api/core/llm/llm_builder.py

@@ -4,9 +4,14 @@ from langchain.callbacks import CallbackManager
 from langchain.llms.fake import FakeListLLM
 
 from core.constant import llm_constant
+from core.llm.error import ProviderTokenNotInitError
+from core.llm.provider.base import BaseProvider
 from core.llm.provider.llm_provider_service import LLMProviderService
+from core.llm.streamable_azure_chat_open_ai import StreamableAzureChatOpenAI
+from core.llm.streamable_azure_open_ai import StreamableAzureOpenAI
 from core.llm.streamable_chat_open_ai import StreamableChatOpenAI
 from core.llm.streamable_open_ai import StreamableOpenAI
+from models.provider import ProviderType
 
 
 class LLMBuilder:
@@ -31,16 +36,23 @@ class LLMBuilder:
         if model_name == 'fake':
             return FakeListLLM(responses=[])
 
+        provider = cls.get_default_provider(tenant_id)
+
         mode = cls.get_mode_by_model(model_name)
         if mode == 'chat':
-            # llm_cls = StreamableAzureChatOpenAI
-            llm_cls = StreamableChatOpenAI
+            if provider == 'openai':
+                llm_cls = StreamableChatOpenAI
+            else:
+                llm_cls = StreamableAzureChatOpenAI
         elif mode == 'completion':
-            llm_cls = StreamableOpenAI
+            if provider == 'openai':
+                llm_cls = StreamableOpenAI
+            else:
+                llm_cls = StreamableAzureOpenAI
         else:
             raise ValueError(f"model name {model_name} is not supported.")
 
-        model_credentials = cls.get_model_credentials(tenant_id, model_name)
+        model_credentials = cls.get_model_credentials(tenant_id, provider, model_name)
 
         return llm_cls(
             model_name=model_name,
@@ -86,18 +98,31 @@ class LLMBuilder:
             raise ValueError(f"model name {model_name} is not supported.")
 
     @classmethod
-    def get_model_credentials(cls, tenant_id: str, model_name: str) -> dict:
+    def get_model_credentials(cls, tenant_id: str, model_provider: str, model_name: str) -> dict:
         """
         Returns the API credentials for the given tenant_id and model_name, based on the model's provider.
         Raises an exception if the model_name is not found or if the provider is not found.
         """
         if not model_name:
             raise Exception('model name not found')
+        #
+        # if model_name not in llm_constant.models:
+        #     raise Exception('model {} not found'.format(model_name))
 
-        if model_name not in llm_constant.models:
-            raise Exception('model {} not found'.format(model_name))
-
-        model_provider = llm_constant.models[model_name]
+        # model_provider = llm_constant.models[model_name]
 
         provider_service = LLMProviderService(tenant_id=tenant_id, provider_name=model_provider)
         return provider_service.get_credentials(model_name)
+
+    @classmethod
+    def get_default_provider(cls, tenant_id: str) -> str:
+        provider = BaseProvider.get_valid_provider(tenant_id)
+        if not provider:
+            raise ProviderTokenNotInitError()
+
+        if provider.provider_type == ProviderType.SYSTEM.value:
+            provider_name = 'openai'
+        else:
+            provider_name = provider.provider_name
+
+        return provider_name

api/core/llm/provider/azure_provider.py

@@ -1,22 +1,24 @@
 import json
+import logging
 from typing import Optional, Union
 
 import requests
 
 from core.llm.provider.base import BaseProvider
+from core.llm.provider.errors import ValidateFailedError
 from models.provider import ProviderName
 
 
 class AzureProvider(BaseProvider):
-    def get_models(self, model_id: Optional[str] = None) -> list[dict]:
-        credentials = self.get_credentials(model_id)
+    def get_models(self, model_id: Optional[str] = None, credentials: Optional[dict] = None) -> list[dict]:
+        credentials = self.get_credentials(model_id) if not credentials else credentials
         url = "{}/openai/deployments?api-version={}".format(
-            credentials.get('openai_api_base'),
-            credentials.get('openai_api_version')
+            str(credentials.get('openai_api_base')),
+            str(credentials.get('openai_api_version'))
         )
 
         headers = {
-            "api-key": credentials.get('openai_api_key'),
+            "api-key": str(credentials.get('openai_api_key')),
             "content-type": "application/json; charset=utf-8"
         }
 
@@ -29,17 +31,18 @@ class AzureProvider(BaseProvider):
                 'name': '{} ({})'.format(deployment['id'], deployment['model'])
             } for deployment in result['data'] if deployment['status'] == 'succeeded']
         else:
-            # TODO: optimize in future
-            raise Exception('Failed to get deployments from Azure OpenAI. Status code: {}'.format(response.status_code))
+            if response.status_code == 401:
+                raise AzureAuthenticationError()
+            else:
+                raise AzureRequestFailedError('Failed to request Azure OpenAI. Status code: {}'.format(response.status_code))
 
     def get_credentials(self, model_id: Optional[str] = None) -> dict:
         """
         Returns the API credentials for Azure OpenAI as a dictionary.
         """
-        encrypted_config = self.get_provider_api_key(model_id=model_id)
-        config = json.loads(encrypted_config)
+        config = self.get_provider_api_key(model_id=model_id)
         config['openai_api_type'] = 'azure'
-        config['deployment_name'] = model_id
+        config['deployment_name'] = model_id.replace('.', '') if model_id else None
         return config
 
     def get_provider_name(self):
@@ -51,12 +54,11 @@ class AzureProvider(BaseProvider):
         """
         try:
             config = self.get_provider_api_key()
-            config = json.loads(config)
         except:
             config = {
                 'openai_api_type': 'azure',
                 'openai_api_version': '2023-03-15-preview',
-                'openai_api_base': 'https://foo.microsoft.com/bar',
+                'openai_api_base': '',
                 'openai_api_key': ''
             }
 
@@ -65,7 +67,7 @@ class AzureProvider(BaseProvider):
                 config = {
                     'openai_api_type': 'azure',
                     'openai_api_version': '2023-03-15-preview',
-                    'openai_api_base': 'https://foo.microsoft.com/bar',
+                    'openai_api_base': '',
                     'openai_api_key': ''
                 }
 
@@ -76,14 +78,50 @@ class AzureProvider(BaseProvider):
 
     def get_token_type(self):
         # TODO: change to dict when implemented
-        return lambda value: value
+        return dict
 
     def config_validate(self, config: Union[dict | str]):
         """
         Validates the given config.
         """
-        # TODO: implement
-        pass
+        try:
+            if not isinstance(config, dict):
+                raise ValueError('Config must be a object.')
+
+            if 'openai_api_version' not in config:
+                config['openai_api_version'] = '2023-03-15-preview'
+
+            models = self.get_models(credentials=config)
+
+            if not models:
+                raise ValidateFailedError("Please add deployments for 'text-davinci-003', "
+                                          "'gpt-3.5-turbo', 'text-embedding-ada-002' (required) "
+                                          "and 'gpt-4', 'gpt-35-turbo-16k' (optional).")
+
+            fixed_model_ids = [
+                'text-davinci-003',
+                'gpt-35-turbo',
+                'text-embedding-ada-002'
+            ]
+
+            current_model_ids = [model['id'] for model in models]
+
+            missing_model_ids = [fixed_model_id for fixed_model_id in fixed_model_ids if
+                                 fixed_model_id not in current_model_ids]
+
+            if missing_model_ids:
+                raise ValidateFailedError("Please add deployments for '{}'.".format(", ".join(missing_model_ids)))
+        except ValidateFailedError as e:
+            raise e
+        except AzureAuthenticationError:
+            raise ValidateFailedError('Validation failed, please check your API Key.')
+        except (requests.ConnectionError, requests.RequestException):
+            raise ValidateFailedError('Validation failed, please check your API Base Endpoint.')
+        except AzureRequestFailedError as ex:
+            raise ValidateFailedError('Validation failed, error: {}.'.format(str(ex)))
+        except Exception as ex:
+            logging.exception('Azure OpenAI Credentials validation failed')
+            raise ValidateFailedError('Validation failed, error: {}.'.format(str(ex)))
 
     def get_encrypted_token(self, config: Union[dict | str]):
         """
@@ -103,3 +141,11 @@ class AzureProvider(BaseProvider):
         config = json.loads(token)
         config['openai_api_key'] = self.decrypt_token(config['openai_api_key'])
         return config
+
+
+class AzureAuthenticationError(Exception):
+    pass
+
+
+class AzureRequestFailedError(Exception):
+    pass

api/core/llm/provider/base.py

@@ -14,7 +14,7 @@ class BaseProvider(ABC):
     def __init__(self, tenant_id: str):
         self.tenant_id = tenant_id
 
-    def get_provider_api_key(self, model_id: Optional[str] = None, prefer_custom: bool = True) -> str:
+    def get_provider_api_key(self, model_id: Optional[str] = None, prefer_custom: bool = True) -> Union[str | dict]:
         """
         Returns the decrypted API key for the given tenant_id and provider_name.
         If the provider is of type SYSTEM and the quota is exceeded, raises a QuotaExceededError.
@@ -43,23 +43,35 @@ class BaseProvider(ABC):
         Returns the Provider instance for the given tenant_id and provider_name.
         If both CUSTOM and System providers exist, the preferred provider will be returned based on the prefer_custom flag.
         """
-        providers = db.session.query(Provider).filter(
-            Provider.tenant_id == self.tenant_id,
-            Provider.provider_name == self.get_provider_name().value
-        ).order_by(Provider.provider_type.desc() if prefer_custom else Provider.provider_type).all()
+        return BaseProvider.get_valid_provider(self.tenant_id, self.get_provider_name().value, prefer_custom)
+
+    @classmethod
+    def get_valid_provider(cls, tenant_id: str, provider_name: str = None, prefer_custom: bool = False) -> Optional[Provider]:
+        """
+        Returns the Provider instance for the given tenant_id and provider_name.
+        If both CUSTOM and System providers exist, the preferred provider will be returned based on the prefer_custom flag.
+        """
+        query = db.session.query(Provider).filter(
+            Provider.tenant_id == tenant_id
+        )
+
+        if provider_name:
+            query = query.filter(Provider.provider_name == provider_name)
+
+        providers = query.order_by(Provider.provider_type.desc() if prefer_custom else Provider.provider_type).all()
 
         custom_provider = None
         system_provider = None
 
         for provider in providers:
-            if provider.provider_type == ProviderType.CUSTOM.value:
+            if provider.provider_type == ProviderType.CUSTOM.value and provider.is_valid and provider.encrypted_config:
                 custom_provider = provider
-            elif provider.provider_type == ProviderType.SYSTEM.value:
+            elif provider.provider_type == ProviderType.SYSTEM.value and provider.is_valid:
                 system_provider = provider
 
-        if custom_provider and custom_provider.is_valid and custom_provider.encrypted_config:
+        if custom_provider:
             return custom_provider
-        elif system_provider and system_provider.is_valid:
+        elif system_provider:
             return system_provider
         else:
             return None
@@ -80,7 +92,7 @@ class BaseProvider(ABC):
         try:
             config = self.get_provider_api_key()
         except:
-            config = 'THIS-IS-A-MOCK-TOKEN'
+            config = ''
 
         if obfuscated:
             return self.obfuscated_token(config)

api/core/llm/streamable_azure_chat_open_ai.py

@@ -1,12 +1,50 @@
-import requests
 from langchain.schema import BaseMessage, ChatResult, LLMResult
 from langchain.chat_models import AzureChatOpenAI
-from typing import Optional, List
+from typing import Optional, List, Dict, Any
+
+from pydantic import root_validator
 
 from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async
 
 
 class StreamableAzureChatOpenAI(AzureChatOpenAI):
+    @root_validator()
+    def validate_environment(cls, values: Dict) -> Dict:
+        """Validate that api key and python package exists in environment."""
+        try:
+            import openai
+        except ImportError:
+            raise ValueError(
+                "Could not import openai python package. "
+                "Please install it with `pip install openai`."
+            )
+        try:
+            values["client"] = openai.ChatCompletion
+        except AttributeError:
+            raise ValueError(
+                "`openai` has no `ChatCompletion` attribute, this is likely "
+                "due to an old version of the openai package. Try upgrading it "
+                "with `pip install --upgrade openai`."
+            )
+        if values["n"] < 1:
+            raise ValueError("n must be at least 1.")
+        if values["n"] > 1 and values["streaming"]:
+            raise ValueError("n must be 1 when streaming.")
+        return values
+
+    @property
+    def _default_params(self) -> Dict[str, Any]:
+        """Get the default parameters for calling OpenAI API."""
+        return {
+            **super()._default_params,
+            "engine": self.deployment_name,
+            "api_type": self.openai_api_type,
+            "api_base": self.openai_api_base,
+            "api_version": self.openai_api_version,
+            "api_key": self.openai_api_key,
+            "organization": self.openai_organization if self.openai_organization else None,
+        }
+
     def get_messages_tokens(self, messages: List[BaseMessage]) -> int:
         """Get the number of tokens in a list of messages.
 

api/core/llm/streamable_azure_open_ai.py

@@ -0,0 +1,64 @@
+import os
+
+from langchain.llms import AzureOpenAI
+from langchain.schema import LLMResult
+from typing import Optional, List, Dict, Mapping, Any
+
+from pydantic import root_validator
+
+from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async
+
+
+class StreamableAzureOpenAI(AzureOpenAI):
+    openai_api_type: str = "azure"
+    openai_api_version: str = ""
+
+    @root_validator()
+    def validate_environment(cls, values: Dict) -> Dict:
+        """Validate that api key and python package exists in environment."""
+        try:
+            import openai
+
+            values["client"] = openai.Completion
+        except ImportError:
+            raise ValueError(
+                "Could not import openai python package. "
+                "Please install it with `pip install openai`."
+            )
+        if values["streaming"] and values["n"] > 1:
+            raise ValueError("Cannot stream results when n > 1.")
+        if values["streaming"] and values["best_of"] > 1:
+            raise ValueError("Cannot stream results when best_of > 1.")
+        return values
+
+    @property
+    def _invocation_params(self) -> Dict[str, Any]:
+        return {**super()._invocation_params, **{
+            "api_type": self.openai_api_type,
+            "api_base": self.openai_api_base,
+            "api_version": self.openai_api_version,
+            "api_key": self.openai_api_key,
+            "organization": self.openai_organization if self.openai_organization else None,
+        }}
+
+    @property
+    def _identifying_params(self) -> Mapping[str, Any]:
+        return {**super()._identifying_params, **{
+            "api_type": self.openai_api_type,
+            "api_base": self.openai_api_base,
+            "api_version": self.openai_api_version,
+            "api_key": self.openai_api_key,
+            "organization": self.openai_organization if self.openai_organization else None,
+        }}
+
+    @handle_llm_exceptions
+    def generate(
+            self, prompts: List[str], stop: Optional[List[str]] = None
+    ) -> LLMResult:
+        return super().generate(prompts, stop)
+
+    @handle_llm_exceptions_async
+    async def agenerate(
+            self, prompts: List[str], stop: Optional[List[str]] = None
+    ) -> LLMResult:
+        return await super().agenerate(prompts, stop)

api/core/llm/streamable_chat_open_ai.py

@@ -1,12 +1,52 @@
+import os
+
 from langchain.schema import BaseMessage, ChatResult, LLMResult
 from langchain.chat_models import ChatOpenAI
-from typing import Optional, List
+from typing import Optional, List, Dict, Any
+
+from pydantic import root_validator
 
 from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async
 
 
 class StreamableChatOpenAI(ChatOpenAI):
 
+    @root_validator()
+    def validate_environment(cls, values: Dict) -> Dict:
+        """Validate that api key and python package exists in environment."""
+        try:
+            import openai
+        except ImportError:
+            raise ValueError(
+                "Could not import openai python package. "
+                "Please install it with `pip install openai`."
+            )
+        try:
+            values["client"] = openai.ChatCompletion
+        except AttributeError:
+            raise ValueError(
+                "`openai` has no `ChatCompletion` attribute, this is likely "
+                "due to an old version of the openai package. Try upgrading it "
+                "with `pip install --upgrade openai`."
+            )
+        if values["n"] < 1:
+            raise ValueError("n must be at least 1.")
+        if values["n"] > 1 and values["streaming"]:
+            raise ValueError("n must be 1 when streaming.")
+        return values
+
+    @property
+    def _default_params(self) -> Dict[str, Any]:
+        """Get the default parameters for calling OpenAI API."""
+        return {
+            **super()._default_params,
+            "api_type": 'openai',
+            "api_base": os.environ.get("OPENAI_API_BASE", "https://api.openai.com/v1"),
+            "api_version": None,
+            "api_key": self.openai_api_key,
+            "organization": self.openai_organization if self.openai_organization else None,
+        }
+
     def get_messages_tokens(self, messages: List[BaseMessage]) -> int:
         """Get the number of tokens in a list of messages.
 

api/core/llm/streamable_open_ai.py

@@ -1,12 +1,54 @@
+import os
+
 from langchain.schema import LLMResult
-from typing import Optional, List
+from typing import Optional, List, Dict, Any, Mapping
 from langchain import OpenAI
+from pydantic import root_validator
 
 from core.llm.error_handle_wraps import handle_llm_exceptions, handle_llm_exceptions_async
 
 
 class StreamableOpenAI(OpenAI):
 
+    @root_validator()
+    def validate_environment(cls, values: Dict) -> Dict:
+        """Validate that api key and python package exists in environment."""
+        try:
+            import openai
+
+            values["client"] = openai.Completion
+        except ImportError:
+            raise ValueError(
+                "Could not import openai python package. "
+                "Please install it with `pip install openai`."
+            )
+        if values["streaming"] and values["n"] > 1:
+            raise ValueError("Cannot stream results when n > 1.")
+        if values["streaming"] and values["best_of"] > 1:
+            raise ValueError("Cannot stream results when best_of > 1.")
+        return values
+
+    @property
+    def _invocation_params(self) -> Dict[str, Any]:
+        return {**super()._invocation_params, **{
+            "api_type": 'openai',
+            "api_base": os.environ.get("OPENAI_API_BASE", "https://api.openai.com/v1"),
+            "api_version": None,
+            "api_key": self.openai_api_key,
+            "organization": self.openai_organization if self.openai_organization else None,
+        }}
+
+    @property
+    def _identifying_params(self) -> Mapping[str, Any]:
+        return {**super()._identifying_params, **{
+            "api_type": 'openai',
+            "api_base": os.environ.get("OPENAI_API_BASE", "https://api.openai.com/v1"),
+            "api_version": None,
+            "api_key": self.openai_api_key,
+            "organization": self.openai_organization if self.openai_organization else None,
+        }}
+
+
     @handle_llm_exceptions
     def generate(
             self, prompts: List[str], stop: Optional[List[str]] = None

api/core/prompt/output_parser/rule_config_generator.py

@@ -0,0 +1,32 @@
+from typing import Any
+
+from langchain.schema import BaseOutputParser, OutputParserException
+from core.prompt.prompts import RULE_CONFIG_GENERATE_TEMPLATE
+from libs.json_in_md_parser import parse_and_check_json_markdown
+
+
+class RuleConfigGeneratorOutputParser(BaseOutputParser):
+
+    def get_format_instructions(self) -> str:
+        return RULE_CONFIG_GENERATE_TEMPLATE
+
+    def parse(self, text: str) -> Any:
+        try:
+            expected_keys = ["prompt", "variables", "opening_statement"]
+            parsed = parse_and_check_json_markdown(text, expected_keys)
+            if not isinstance(parsed["prompt"], str):
+                raise ValueError("Expected 'prompt' to be a string.")
+            if not isinstance(parsed["variables"], list):
+                raise ValueError(
+                    f"Expected 'variables' to be a list."
+                )
+            if not isinstance(parsed["opening_statement"], str):
+                raise ValueError(
+                    f"Expected 'opening_statement' to be a str."
+                )
+            return parsed
+        except Exception as e:
+            raise OutputParserException(
+                f"Parsing text\n{text}\n of rule config generator raised following error:\n{e}"
+            )
+

api/core/prompt/prompt_builder.py

@@ -32,6 +32,6 @@ class PromptBuilder:
 
     @classmethod
     def process_template(cls, template: str):
-        processed_template = re.sub(r'\{(.+?)\}', r'\1', template)
-        processed_template = re.sub(r'\{\{(.+?)\}\}', r'{\1}', processed_template)
+        processed_template = re.sub(r'\{([a-zA-Z_]\w+?)\}', r'\1', template)
+        processed_template = re.sub(r'\{\{([a-zA-Z_]\w+?)\}\}', r'{\1}', processed_template)
         return processed_template

api/core/prompt/prompts.py

@@ -61,3 +61,60 @@ QUERY_KEYWORD_EXTRACT_TEMPLATE_TMPL = (
 QUERY_KEYWORD_EXTRACT_TEMPLATE = QueryKeywordExtractPrompt(
     QUERY_KEYWORD_EXTRACT_TEMPLATE_TMPL
 )
+
+RULE_CONFIG_GENERATE_TEMPLATE = """Given MY INTENDED AUDIENCES and HOPING TO SOLVE using a language model, please select \
+the model prompt that best suits the input. 
+You will be provided with the prompt, variables, and an opening statement. 
+Only the content enclosed in double curly braces, such as {{variable}}, in the prompt can be considered as a variable; \
+otherwise, it cannot exist as a variable in the variables.
+If you believe revising the original input will result in a better response from the language model, you may \
+suggest revisions.
+
+<< FORMATTING >>
+Return a markdown code snippet with a JSON object formatted to look like, \
+no any other string out of markdown code snippet:
+```json
+{{{{
+    "prompt": string \\ generated prompt
+    "variables": list of string \\ variables
+    "opening_statement": string \\ an opening statement to guide users on how to ask questions with generated prompt \
+and fill in variables, with a welcome sentence, and keep TLDR.
+}}}}
+```
+
+<< EXAMPLES >>
+[EXAMPLE A]
+```json
+{
+  "prompt": "Write a letter about love",
+  "variables": [],
+  "opening_statement": "Hi! I'm your love letter writer AI."
+}
+```
+
+[EXAMPLE B]
+```json
+{
+  "prompt": "Translate from {{lanA}} to {{lanB}}",
+  "variables": ["lanA", "lanB"],
+  "opening_statement": "Welcome to use translate app"
+}
+```
+
+[EXAMPLE C]
+```json
+{
+  "prompt": "Write a story about {{topic}}",
+  "variables": ["topic"],
+  "opening_statement": "I'm your story writer"
+}
+```
+
+<< MY INTENDED AUDIENCES >>
+{audiences}
+
+<< HOPING TO SOLVE >>
+{hoping_to_solve}
+
+<< OUTPUT >>
+"""

api/core/tool/dataset_tool_builder.py

@@ -10,24 +10,14 @@ from core.index.keyword_table_index import KeywordTableIndex
 from core.index.vector_index import VectorIndex
 from core.prompt.prompts import QUERY_KEYWORD_EXTRACT_TEMPLATE
 from core.tool.llama_index_tool import EnhanceLlamaIndexTool
-from extensions.ext_database import db
 from models.dataset import Dataset
 
 
 class DatasetToolBuilder:
     @classmethod
-    def build_dataset_tool(cls, tenant_id: str, dataset_id: str,
+    def build_dataset_tool(cls, dataset: Dataset,
                            response_mode: str = "no_synthesizer",
                            callback_handler: Optional[DatasetToolCallbackHandler] = None):
-        # get dataset from dataset id
-        dataset = db.session.query(Dataset).filter(
-            Dataset.tenant_id == tenant_id,
-            Dataset.id == dataset_id
-        ).first()
-
-        if not dataset:
-            return None
-
         if dataset.indexing_technique == "economy":
             # use keyword table query
             index = KeywordTableIndex(dataset=dataset).query_index
@@ -65,7 +55,7 @@ class DatasetToolBuilder:
 
         index_tool_config = IndexToolConfig(
             index=index,
-            name=f"dataset-{dataset_id}",
+            name=f"dataset-{dataset.id}",
             description=description,
             index_query_kwargs=query_kwargs,
             tool_kwargs={
@@ -75,7 +65,7 @@ class DatasetToolBuilder:
             # return_direct: Whether to return LLM results directly or process the output data with an Output Parser
         )
 
-        index_callback_handler = DatasetIndexToolCallbackHandler(dataset_id=dataset_id)
+        index_callback_handler = DatasetIndexToolCallbackHandler(dataset_id=dataset.id)
 
         return EnhanceLlamaIndexTool.from_tool_config(
             tool_config=index_tool_config,

api/core/vector_store/vector_store.py

@@ -27,7 +27,8 @@ class VectorStore:
             self._client = WeaviateVectorStoreClient(
                 endpoint=app.config['WEAVIATE_ENDPOINT'],
                 api_key=app.config['WEAVIATE_API_KEY'],
-                grpc_enabled=app.config['WEAVIATE_GRPC_ENABLED']
+                grpc_enabled=app.config['WEAVIATE_GRPC_ENABLED'],
+                batch_size=app.config['WEAVIATE_BATCH_SIZE']
             )
         elif self._vector_store == 'qdrant':
             self._client = QdrantVectorStoreClient(

api/core/vector_store/weaviate_vector_store_client.py

@@ -18,21 +18,33 @@ from llama_index.readers.weaviate.utils import (
 
 class WeaviateVectorStoreClient(BaseVectorStoreClient):
 
-    def __init__(self, endpoint: str, api_key: str, grpc_enabled: bool):
-        self._client = self.init_from_config(endpoint, api_key, grpc_enabled)
+    def __init__(self, endpoint: str, api_key: str, grpc_enabled: bool, batch_size: int):
+        self._client = self.init_from_config(endpoint, api_key, grpc_enabled, batch_size)
 
-    def init_from_config(self, endpoint: str, api_key: str, grpc_enabled: bool):
+    def init_from_config(self, endpoint: str, api_key: str, grpc_enabled: bool, batch_size: int):
         auth_config = weaviate.auth.AuthApiKey(api_key=api_key)
 
         weaviate.connect.connection.has_grpc = grpc_enabled
 
-        return weaviate.Client(
+        client = weaviate.Client(
             url=endpoint,
             auth_client_secret=auth_config,
-            timeout_config=(5, 15),
+            timeout_config=(5, 60),
             startup_period=None
         )
 
+        client.batch.configure(
+            # `batch_size` takes an `int` value to enable auto-batching
+            # (`None` is used for manual batching)
+            batch_size=batch_size,
+            # dynamically update the `batch_size` based on import speed
+            dynamic=True,
+            # `timeout_retries` takes an `int` value to retry on time outs
+            timeout_retries=3,
+        )
+
+        return client
+
     def get_index(self, service_context: ServiceContext, config: dict) -> GPTVectorStoreIndex:
         index_struct = WeaviateIndexDict()
 

api/extensions/ext_celery.py

@@ -15,9 +15,24 @@ def init_app(app: Flask) -> Celery:
         backend=app.config["CELERY_BACKEND"],
         task_ignore_result=True,
     )
+    
+    # Add SSL options to the Celery configuration
+    ssl_options = {
+        "ssl_cert_reqs": None,
+        "ssl_ca_certs": None,
+        "ssl_certfile": None,
+        "ssl_keyfile": None,
+    }
+
     celery_app.conf.update(
         result_backend=app.config["CELERY_RESULT_BACKEND"],
     )
+
+    if app.config["BROKER_USE_SSL"]:
+        celery_app.conf.update(
+            broker_use_ssl=ssl_options,  # Add the SSL options to the broker configuration
+        )
+        
     celery_app.set_default()
     app.extensions["celery"] = celery_app
     return celery_app

api/extensions/ext_redis.py

@@ -1,18 +1,23 @@
 import redis
-
+from redis.connection import SSLConnection, Connection
 
 redis_client = redis.Redis()
 
 
 def init_app(app):
+    connection_class = Connection
+    if app.config.get('REDIS_USE_SSL', False):
+        connection_class = SSLConnection
+
     redis_client.connection_pool = redis.ConnectionPool(**{
         'host': app.config.get('REDIS_HOST', 'localhost'),
         'port': app.config.get('REDIS_PORT', 6379),
+        'username': app.config.get('REDIS_USERNAME', None),
         'password': app.config.get('REDIS_PASSWORD', None),
         'db': app.config.get('REDIS_DB', 0),
         'encoding': 'utf-8',
         'encoding_errors': 'strict',
         'decode_responses': False
-    })
+    }, connection_class=connection_class)
 
     app.extensions['redis'] = redis_client

api/extensions/ext_session.py

@@ -1,4 +1,5 @@
 import redis
+from redis.connection import SSLConnection, Connection
 from flask import request
 from flask_session import Session, SqlAlchemySessionInterface, RedisSessionInterface
 from flask_session.sessions import total_seconds
@@ -23,16 +24,21 @@ def init_app(app):
     if session_type == 'sqlalchemy':
         app.session_interface = sqlalchemy_session_interface
     elif session_type == 'redis':
+        connection_class = Connection
+        if app.config.get('SESSION_REDIS_USE_SSL', False):
+            connection_class = SSLConnection
+
         sess_redis_client = redis.Redis()
         sess_redis_client.connection_pool = redis.ConnectionPool(**{
             'host': app.config.get('SESSION_REDIS_HOST', 'localhost'),
             'port': app.config.get('SESSION_REDIS_PORT', 6379),
+            'username': app.config.get('SESSION_REDIS_USERNAME', None),
             'password': app.config.get('SESSION_REDIS_PASSWORD', None),
             'db': app.config.get('SESSION_REDIS_DB', 2),
             'encoding': 'utf-8',
             'encoding_errors': 'strict',
             'decode_responses': False
-        })
+        }, connection_class=connection_class)
 
         app.extensions['session_redis'] = sess_redis_client
 

api/libs/helper.py

@@ -21,7 +21,7 @@ class TimestampField(fields.Raw):
 
 def email(email):
     # Define a regex pattern for email addresses
-    pattern = r"^[a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+$"
+    pattern = r"^[\w\.-]+@([\w-]+\.)+[\w-]{2,}$"
     # Check if the email matches the pattern
     if re.match(pattern, email) is not None:
         return email

api/libs/json_in_md_parser.py

@@ -0,0 +1,44 @@
+import json
+from typing import List
+
+from langchain.schema import OutputParserException
+
+
+def parse_json_markdown(json_string: str) -> dict:
+    # Remove the triple backticks if present
+    json_string = json_string.strip()
+    start_index = json_string.find("```json")
+    end_index = json_string.find("```", start_index + len("```json"))
+
+    if start_index != -1 and end_index != -1:
+        extracted_content = json_string[start_index + len("```json"):end_index].strip()
+
+        # Parse the JSON string into a Python dictionary
+        parsed = json.loads(extracted_content)
+    elif start_index != -1 and end_index == -1 and json_string.endswith("``"):
+        end_index = json_string.find("``", start_index + len("```json"))
+        extracted_content = json_string[start_index + len("```json"):end_index].strip()
+
+        # Parse the JSON string into a Python dictionary
+        parsed = json.loads(extracted_content)
+    elif json_string.startswith("{"):
+        # Parse the JSON string into a Python dictionary
+        parsed = json.loads(json_string)
+    else:
+        raise Exception("Could not find JSON block in the output.")
+
+    return parsed
+
+
+def parse_and_check_json_markdown(text: str, expected_keys: List[str]) -> dict:
+    try:
+        json_obj = parse_json_markdown(text)
+    except json.JSONDecodeError as e:
+        raise OutputParserException(f"Got invalid JSON object. Error: {e}")
+    for key in expected_keys:
+        if key not in json_obj:
+            raise OutputParserException(
+                f"Got invalid return object. Expected key `{key}` "
+                f"to be present, but got {json_obj}"
+            )
+    return json_obj

api/libs/oauth.py

@@ -1,7 +1,12 @@
+import json
 import urllib.parse
 from dataclasses import dataclass
 
 import requests
+from flask_login import current_user
+
+from extensions.ext_database import db
+from models.source import DataSourceBinding
 
 
 @dataclass
@@ -134,3 +139,5 @@ class GoogleOAuth(OAuth):
             name=None,
             email=raw_info['email']
         )
+
+

api/libs/oauth_data_source.py

@@ -0,0 +1,312 @@
+import json
+import urllib.parse
+
+import requests
+from flask_login import current_user
+
+from extensions.ext_database import db
+from models.source import DataSourceBinding
+
+
+class OAuthDataSource:
+    def __init__(self, client_id: str, client_secret: str, redirect_uri: str):
+        self.client_id = client_id
+        self.client_secret = client_secret
+        self.redirect_uri = redirect_uri
+
+    def get_authorization_url(self):
+        raise NotImplementedError()
+
+    def get_access_token(self, code: str):
+        raise NotImplementedError()
+
+
+class NotionOAuth(OAuthDataSource):
+    _AUTH_URL = 'https://api.notion.com/v1/oauth/authorize'
+    _TOKEN_URL = 'https://api.notion.com/v1/oauth/token'
+    _NOTION_PAGE_SEARCH = "https://api.notion.com/v1/search"
+    _NOTION_BLOCK_SEARCH = "https://api.notion.com/v1/blocks"
+    _NOTION_BOT_USER = "https://api.notion.com/v1/users/me"
+
+    def get_authorization_url(self):
+        params = {
+            'client_id': self.client_id,
+            'response_type': 'code',
+            'redirect_uri': self.redirect_uri,
+            'owner': 'user'
+        }
+        return f"{self._AUTH_URL}?{urllib.parse.urlencode(params)}"
+
+    def get_access_token(self, code: str):
+        data = {
+            'code': code,
+            'grant_type': 'authorization_code',
+            'redirect_uri': self.redirect_uri
+        }
+        headers = {'Accept': 'application/json'}
+        auth = (self.client_id, self.client_secret)
+        response = requests.post(self._TOKEN_URL, data=data, auth=auth, headers=headers)
+
+        response_json = response.json()
+        access_token = response_json.get('access_token')
+        if not access_token:
+            raise ValueError(f"Error in Notion OAuth: {response_json}")
+        workspace_name = response_json.get('workspace_name')
+        workspace_icon = response_json.get('workspace_icon')
+        workspace_id = response_json.get('workspace_id')
+        # get all authorized pages
+        pages = self.get_authorized_pages(access_token)
+        source_info = {
+            'workspace_name': workspace_name,
+            'workspace_icon': workspace_icon,
+            'workspace_id': workspace_id,
+            'pages': pages,
+            'total': len(pages)
+        }
+        # save data source binding
+        data_source_binding = DataSourceBinding.query.filter(
+            db.and_(
+                DataSourceBinding.tenant_id == current_user.current_tenant_id,
+                DataSourceBinding.provider == 'notion',
+                DataSourceBinding.access_token == access_token
+            )
+        ).first()
+        if data_source_binding:
+            data_source_binding.source_info = source_info
+            data_source_binding.disabled = False
+            db.session.commit()
+        else:
+            new_data_source_binding = DataSourceBinding(
+                tenant_id=current_user.current_tenant_id,
+                access_token=access_token,
+                source_info=source_info,
+                provider='notion'
+            )
+            db.session.add(new_data_source_binding)
+            db.session.commit()
+
+    def save_internal_access_token(self, access_token: str):
+        workspace_name = self.notion_workspace_name(access_token)
+        workspace_icon = None
+        workspace_id = current_user.current_tenant_id
+        # get all authorized pages
+        pages = self.get_authorized_pages(access_token)
+        source_info = {
+            'workspace_name': workspace_name,
+            'workspace_icon': workspace_icon,
+            'workspace_id': workspace_id,
+            'pages': pages,
+            'total': len(pages)
+        }
+        # save data source binding
+        data_source_binding = DataSourceBinding.query.filter(
+            db.and_(
+                DataSourceBinding.tenant_id == current_user.current_tenant_id,
+                DataSourceBinding.provider == 'notion',
+                DataSourceBinding.access_token == access_token
+            )
+        ).first()
+        if data_source_binding:
+            data_source_binding.source_info = source_info
+            data_source_binding.disabled = False
+            db.session.commit()
+        else:
+            new_data_source_binding = DataSourceBinding(
+                tenant_id=current_user.current_tenant_id,
+                access_token=access_token,
+                source_info=source_info,
+                provider='notion'
+            )
+            db.session.add(new_data_source_binding)
+            db.session.commit()
+
+    def sync_data_source(self, binding_id: str):
+        # save data source binding
+        data_source_binding = DataSourceBinding.query.filter(
+            db.and_(
+                DataSourceBinding.tenant_id == current_user.current_tenant_id,
+                DataSourceBinding.provider == 'notion',
+                DataSourceBinding.id == binding_id,
+                DataSourceBinding.disabled == False
+            )
+        ).first()
+        if data_source_binding:
+            # get all authorized pages
+            pages = self.get_authorized_pages(data_source_binding.access_token)
+            source_info = data_source_binding.source_info
+            new_source_info = {
+                'workspace_name': source_info['workspace_name'],
+                'workspace_icon': source_info['workspace_icon'],
+                'workspace_id': source_info['workspace_id'],
+                'pages': pages,
+                'total': len(pages)
+            }
+            data_source_binding.source_info = new_source_info
+            data_source_binding.disabled = False
+            db.session.commit()
+        else:
+            raise ValueError('Data source binding not found')
+
+    def get_authorized_pages(self, access_token: str):
+        pages = []
+        page_results = self.notion_page_search(access_token)
+        database_results = self.notion_database_search(access_token)
+        # get page detail
+        for page_result in page_results:
+            page_id = page_result['id']
+            if 'Name' in page_result['properties']:
+                if len(page_result['properties']['Name']['title']) > 0:
+                    page_name = page_result['properties']['Name']['title'][0]['plain_text']
+                else:
+                    page_name = 'Untitled'
+            elif 'title' in page_result['properties']:
+                if len(page_result['properties']['title']['title']) > 0:
+                    page_name = page_result['properties']['title']['title'][0]['plain_text']
+                else:
+                    page_name = 'Untitled'
+            elif 'Title' in page_result['properties']:
+                if len(page_result['properties']['Title']['title']) > 0:
+                    page_name = page_result['properties']['Title']['title'][0]['plain_text']
+                else:
+                    page_name = 'Untitled'
+            else:
+                page_name = 'Untitled'
+            page_icon = page_result['icon']
+            if page_icon:
+                icon_type = page_icon['type']
+                if icon_type == 'external' or icon_type == 'file':
+                    url = page_icon[icon_type]['url']
+                    icon = {
+                        'type': 'url',
+                        'url': url if url.startswith('http') else f'https://www.notion.so{url}'
+                    }
+                else:
+                    icon = {
+                        'type': 'emoji',
+                        'emoji': page_icon[icon_type]
+                    }
+            else:
+                icon = None
+            parent = page_result['parent']
+            parent_type = parent['type']
+            if parent_type == 'block_id':
+                parent_id = self.notion_block_parent_page_id(access_token, parent[parent_type])
+            elif parent_type == 'workspace':
+                parent_id = 'root'
+            else:
+                parent_id = parent[parent_type]
+            page = {
+                'page_id': page_id,
+                'page_name': page_name,
+                'page_icon': icon,
+                'parent_id': parent_id,
+                'type': 'page'
+            }
+            pages.append(page)
+            # get database detail
+        for database_result in database_results:
+            page_id = database_result['id']
+            if len(database_result['title']) > 0:
+                page_name = database_result['title'][0]['plain_text']
+            else:
+                page_name = 'Untitled'
+            page_icon = database_result['icon']
+            if page_icon:
+                icon_type = page_icon['type']
+                if icon_type == 'external' or icon_type == 'file':
+                    url = page_icon[icon_type]['url']
+                    icon = {
+                        'type': 'url',
+                        'url': url if url.startswith('http') else f'https://www.notion.so{url}'
+                    }
+                else:
+                    icon = {
+                        'type': icon_type,
+                        icon_type: page_icon[icon_type]
+                    }
+            else:
+                icon = None
+            parent = database_result['parent']
+            parent_type = parent['type']
+            if parent_type == 'block_id':
+                parent_id = self.notion_block_parent_page_id(access_token, parent[parent_type])
+            elif parent_type == 'workspace':
+                parent_id = 'root'
+            else:
+                parent_id = parent[parent_type]
+            page = {
+                'page_id': page_id,
+                'page_name': page_name,
+                'page_icon': icon,
+                'parent_id': parent_id,
+                'type': 'database'
+            }
+            pages.append(page)
+        return pages
+
+    def notion_page_search(self, access_token: str):
+        data = {
+            'filter': {
+                "value": "page",
+                "property": "object"
+            }
+        }
+        headers = {
+            'Content-Type': 'application/json',
+            'Authorization': f"Bearer {access_token}",
+            'Notion-Version': '2022-06-28',
+        }
+        response = requests.post(url=self._NOTION_PAGE_SEARCH, json=data, headers=headers)
+        response_json = response.json()
+        if 'results' in response_json:
+            results = response_json['results']
+        else:
+            results = []
+        return results
+
+    def notion_block_parent_page_id(self, access_token: str, block_id: str):
+        headers = {
+            'Authorization': f"Bearer {access_token}",
+            'Notion-Version': '2022-06-28',
+        }
+        response = requests.get(url=f'{self._NOTION_BLOCK_SEARCH}/{block_id}', headers=headers)
+        response_json = response.json()
+        parent = response_json['parent']
+        parent_type = parent['type']
+        if parent_type == 'block_id':
+            return self.notion_block_parent_page_id(access_token, parent[parent_type])
+        return parent[parent_type]
+
+    def notion_workspace_name(self, access_token: str):
+        headers = {
+            'Authorization': f"Bearer {access_token}",
+            'Notion-Version': '2022-06-28',
+        }
+        response = requests.get(url=self._NOTION_BOT_USER, headers=headers)
+        response_json = response.json()
+        if 'object' in response_json and response_json['object'] == 'user':
+            user_type = response_json['type']
+            user_info = response_json[user_type]
+            if 'workspace_name' in user_info:
+                return user_info['workspace_name']
+        return 'workspace'
+
+    def notion_database_search(self, access_token: str):
+        data = {
+            'filter': {
+                "value": "database",
+                "property": "object"
+            }
+        }
+        headers = {
+            'Content-Type': 'application/json',
+            'Authorization': f"Bearer {access_token}",
+            'Notion-Version': '2022-06-28',
+        }
+        response = requests.post(url=self._NOTION_PAGE_SEARCH, json=data, headers=headers)
+        response_json = response.json()
+        if 'results' in response_json:
+            results = response_json['results']
+        else:
+            results = []
+        return results

api/migrations/versions/614f77cecc48_add_last_active_at.py

@@ -0,0 +1,32 @@
+"""add last active at
+
+Revision ID: 614f77cecc48
+Revises: a45f4dfde53b
+Create Date: 2023-06-15 13:33:00.357467
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = '614f77cecc48'
+down_revision = 'a45f4dfde53b'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('accounts', schema=None) as batch_op:
+        batch_op.add_column(sa.Column('last_active_at', sa.DateTime(), server_default=sa.text('CURRENT_TIMESTAMP(0)'), nullable=False))
+
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('accounts', schema=None) as batch_op:
+        batch_op.drop_column('last_active_at')
+
+    # ### end Alembic commands ###

api/migrations/versions/64b051264f32_init.py

@@ -18,6 +18,8 @@ depends_on = None
 
 def upgrade():
     # ### commands auto generated by Alembic - please adjust! ###
+    op.execute('CREATE EXTENSION IF NOT EXISTS "uuid-ossp";')
+
     op.create_table('account_integrates',
     sa.Column('id', postgresql.UUID(), server_default=sa.text('uuid_generate_v4()'), nullable=False),
     sa.Column('account_id', postgresql.UUID(), nullable=False),
@@ -790,4 +792,6 @@ def downgrade():
 
     op.drop_table('accounts')
     op.drop_table('account_integrates')
+
+    op.execute('DROP EXTENSION IF EXISTS "uuid-ossp";')
     # ### end Alembic commands ###

api/migrations/versions/9f4e3427ea84_add_created_by_role.py

@@ -0,0 +1,46 @@
+"""add created by role
+
+Revision ID: 9f4e3427ea84
+Revises: 64b051264f32
+Create Date: 2023-05-17 17:29:01.060435
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = '9f4e3427ea84'
+down_revision = '64b051264f32'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('pinned_conversations', schema=None) as batch_op:
+        batch_op.add_column(sa.Column('created_by_role', sa.String(length=255), server_default=sa.text("'end_user'::character varying"), nullable=False))
+        batch_op.drop_index('pinned_conversation_conversation_idx')
+        batch_op.create_index('pinned_conversation_conversation_idx', ['app_id', 'conversation_id', 'created_by_role', 'created_by'], unique=False)
+
+    with op.batch_alter_table('saved_messages', schema=None) as batch_op:
+        batch_op.add_column(sa.Column('created_by_role', sa.String(length=255), server_default=sa.text("'end_user'::character varying"), nullable=False))
+        batch_op.drop_index('saved_message_message_idx')
+        batch_op.create_index('saved_message_message_idx', ['app_id', 'message_id', 'created_by_role', 'created_by'], unique=False)
+
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('saved_messages', schema=None) as batch_op:
+        batch_op.drop_index('saved_message_message_idx')
+        batch_op.create_index('saved_message_message_idx', ['app_id', 'message_id', 'created_by'], unique=False)
+        batch_op.drop_column('created_by_role')
+
+    with op.batch_alter_table('pinned_conversations', schema=None) as batch_op:
+        batch_op.drop_index('pinned_conversation_conversation_idx')
+        batch_op.create_index('pinned_conversation_conversation_idx', ['app_id', 'conversation_id', 'created_by'], unique=False)
+        batch_op.drop_column('created_by_role')
+
+    # ### end Alembic commands ###

api/migrations/versions/a45f4dfde53b_add_language_to_recommend_apps.py

@@ -0,0 +1,36 @@
+"""add language to recommend apps
+
+Revision ID: a45f4dfde53b
+Revises: 9f4e3427ea84
+Create Date: 2023-05-25 17:50:32.052335
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = 'a45f4dfde53b'
+down_revision = '9f4e3427ea84'
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('recommended_apps', schema=None) as batch_op:
+        batch_op.add_column(sa.Column('language', sa.String(length=255), server_default=sa.text("'en-US'::character varying"), nullable=False))
+        batch_op.drop_index('recommended_app_is_listed_idx')
+        batch_op.create_index('recommended_app_is_listed_idx', ['is_listed', 'language'], unique=False)
+
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    with op.batch_alter_table('recommended_apps', schema=None) as batch_op:
+        batch_op.drop_index('recommended_app_is_listed_idx')
+        batch_op.create_index('recommended_app_is_listed_idx', ['is_listed'], unique=False)
+        batch_op.drop_column('language')
+
+    # ### end Alembic commands ###