Merge branch 'fix/agent-external-knowledge-retrieval' into fix/external-knowledge-retrieval-issues

.github/workflows/api-tests.yml

@@ -39,7 +39,7 @@ jobs:
             api/pyproject.toml
             api/poetry.lock
 
-      - name: Poetry check
+      - name: Check Poetry lockfile
         run: |
           poetry check -C api --lock
           poetry show -C api
@@ -47,6 +47,9 @@ jobs:
       - name: Install dependencies
         run: poetry install -C api --with dev
 
+      - name: Check dependencies in pyproject.toml
+        run: poetry run -C api bash dev/pytest/pytest_artifacts.sh
+
       - name: Run Unit tests
         run: poetry run -C api bash dev/pytest/pytest_unit_tests.sh
 

README.md

@@ -17,7 +17,7 @@
             alt="chat on Discord"></a>
     <a href="https://twitter.com/intent/follow?screen_name=dify_ai" target="_blank">
         <img src="https://img.shields.io/twitter/follow/dify_ai?logo=X&color=%20%23f5f5f5"
-            alt="follow on Twitter"></a>
+            alt="follow on X(Twitter)"></a>
     <a href="https://hub.docker.com/u/langgenius" target="_blank">
         <img alt="Docker Pulls" src="https://img.shields.io/docker/pulls/langgenius/dify-web?labelColor=%20%23FDB062&color=%20%23f79009"></a>
     <a href="https://github.com/langgenius/dify/graphs/commit-activity" target="_blank">
@@ -196,10 +196,14 @@ If you'd like to configure a highly-available setup, there are community-contrib
 
 #### Using Terraform for Deployment
 
+Deploy Dify to Cloud Platform with a single click using [terraform](https://www.terraform.io/)
+
 ##### Azure Global
-Deploy Dify to Azure with a single click using [terraform](https://www.terraform.io/).
 - [Azure Terraform by @nikawang](https://github.com/nikawang/dify-azure-terraform)
 
+##### Google Cloud
+- [Google Cloud Terraform by @sotazum](https://github.com/DeNA/dify-google-cloud-terraform)
+
 ## Contributing
 
 For those who'd like to contribute code, see our [Contribution Guide](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md). 
@@ -219,7 +223,7 @@ At the same time, please consider supporting Dify by sharing it on social media
 * [Github Discussion](https://github.com/langgenius/dify/discussions). Best for: sharing feedback and asking questions.
 * [GitHub Issues](https://github.com/langgenius/dify/issues). Best for: bugs you encounter using Dify.AI, and feature proposals. See our [Contribution Guide](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md).
 * [Discord](https://discord.gg/FngNHpbcY7). Best for: sharing your applications and hanging out with the community.
-* [Twitter](https://twitter.com/dify_ai). Best for: sharing your applications and hanging out with the community.
+* [X(Twitter)](https://twitter.com/dify_ai). Best for: sharing your applications and hanging out with the community.
 
 ## Star history
 

README_AR.md

@@ -17,7 +17,7 @@
             alt="chat on Discord"></a>
     <a href="https://twitter.com/intent/follow?screen_name=dify_ai" target="_blank">
         <img src="https://img.shields.io/twitter/follow/dify_ai?logo=X&color=%20%23f5f5f5"
-            alt="follow on Twitter"></a>
+            alt="follow on X(Twitter)"></a>
     <a href="https://hub.docker.com/u/langgenius" target="_blank">
         <img alt="Docker Pulls" src="https://img.shields.io/docker/pulls/langgenius/dify-web?labelColor=%20%23FDB062&color=%20%23f79009"></a>
     <a href="https://github.com/langgenius/dify/graphs/commit-activity" target="_blank">
@@ -179,10 +179,13 @@ docker compose up -d
 
 #### استخدام Terraform للتوزيع
 
+انشر Dify إلى منصة السحابة بنقرة واحدة باستخدام [terraform](https://www.terraform.io/)
+
 ##### Azure Global
-استخدم [terraform](https://www.terraform.io/) لنشر Dify على Azure بنقرة واحدة.
 - [Azure Terraform بواسطة @nikawang](https://github.com/nikawang/dify-azure-terraform)
 
+##### Google Cloud
+- [Google Cloud Terraform بواسطة @sotazum](https://github.com/DeNA/dify-google-cloud-terraform)
 
 ## المساهمة
 

README_CN.md

@@ -17,7 +17,7 @@
             alt="chat on Discord"></a>
     <a href="https://twitter.com/intent/follow?screen_name=dify_ai" target="_blank">
         <img src="https://img.shields.io/twitter/follow/dify_ai?logo=X&color=%20%23f5f5f5"
-            alt="follow on Twitter"></a>
+            alt="follow on X(Twitter)"></a>
     <a href="https://hub.docker.com/u/langgenius" target="_blank">
         <img alt="Docker Pulls" src="https://img.shields.io/docker/pulls/langgenius/dify-web?labelColor=%20%23FDB062&color=%20%23f79009"></a>
     <a href="https://github.com/langgenius/dify/graphs/commit-activity" target="_blank">
@@ -202,10 +202,14 @@ docker compose up -d
 
 #### 使用 Terraform 部署
 
+使用 [terraform](https://www.terraform.io/) 一键将 Dify 部署到云平台
+
 ##### Azure Global
-使用 [terraform](https://www.terraform.io/) 一键部署 Dify 到 Azure。
 - [Azure Terraform by @nikawang](https://github.com/nikawang/dify-azure-terraform)
 
+##### Google Cloud
+- [Google Cloud Terraform by @sotazum](https://github.com/DeNA/dify-google-cloud-terraform)
+
 ## Star History
 
 [![Star History Chart](https://api.star-history.com/svg?repos=langgenius/dify&type=Date)](https://star-history.com/#langgenius/dify&Date)
@@ -232,7 +236,7 @@ docker compose up -d
 - [GitHub Issues](https://github.com/langgenius/dify/issues)。👉：使用 Dify.AI 时遇到的错误和问题，请参阅[贡献指南](CONTRIBUTING.md)。
 - [电子邮件支持](mailto:hello@dify.ai?subject=[GitHub]Questions%20About%20Dify)。👉：关于使用 Dify.AI 的问题。
 - [Discord](https://discord.gg/FngNHpbcY7)。👉：分享您的应用程序并与社区交流。
-- [Twitter](https://twitter.com/dify_ai)。👉：分享您的应用程序并与社区交流。
+- [X(Twitter)](https://twitter.com/dify_ai)。👉：分享您的应用程序并与社区交流。
 - [商业许可](mailto:business@dify.ai?subject=[GitHub]Business%20License%20Inquiry)。👉：有关商业用途许可 Dify.AI 的商业咨询。
  - [微信]() 👉：扫描下方二维码，添加微信好友，备注 Dify，我们将邀请您加入 Dify 社区。  
 <img src="./images/wechat.png" alt="wechat" width="100"/>

README_ES.md

@@ -17,7 +17,7 @@
             alt="chat en Discord"></a>
     <a href="https://twitter.com/intent/follow?screen_name=dify_ai" target="_blank">
         <img src="https://img.shields.io/twitter/follow/dify_ai?logo=X&color=%20%23f5f5f5"
-            alt="seguir en Twitter"></a>
+            alt="seguir en X(Twitter)"></a>
     <a href="https://hub.docker.com/u/langgenius" target="_blank">
         <img alt="Descargas de Docker" src="https://img.shields.io/docker/pulls/langgenius/dify-web?labelColor=%20%23FDB062&color=%20%23f79009"></a>
     <a href="https://github.com/langgenius/dify/graphs/commit-activity" target="_blank">
@@ -204,10 +204,13 @@ Si desea configurar una configuración de alta disponibilidad, la comunidad prop
 
 #### Uso de Terraform para el despliegue
 
+Despliega Dify en una plataforma en la nube con un solo clic utilizando [terraform](https://www.terraform.io/)
+
 ##### Azure Global
-Utiliza [terraform](https://www.terraform.io/) para desplegar Dify en Azure con un solo clic.
 - [Azure Terraform por @nikawang](https://github.com/nikawang/dify-azure-terraform)
 
+##### Google Cloud
+- [Google Cloud Terraform por @sotazum](https://github.com/DeNA/dify-google-cloud-terraform)
 
 ## Contribuir
 
@@ -228,7 +231,7 @@ Al mismo tiempo, considera apoyar a Dify compartiéndolo en redes sociales y en
 * [Discusión en GitHub](https://github.com/langgenius/dify/discussions). Lo mejor para: compartir comentarios y hacer preguntas.
 * [Reporte de problemas en GitHub](https://github.com/langgenius/dify/issues). Lo mejor para: errores que encuentres usando Dify.AI y propuestas de características. Consulta nuestra [Guía de contribución](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md).
 * [Discord](https://discord.gg/FngNHpbcY7). Lo mejor para: compartir tus aplicaciones y pasar el rato con la comunidad.
-* [Twitter](https://twitter.com/dify_ai). Lo mejor para: compartir tus aplicaciones y pasar el rato con la comunidad.
+* [X(Twitter)](https://twitter.com/dify_ai). Lo mejor para: compartir tus aplicaciones y pasar el rato con la comunidad.
 
 ## Historial de Estrellas
 

README_FR.md

@@ -17,7 +17,7 @@
             alt="chat sur Discord"></a>
     <a href="https://twitter.com/intent/follow?screen_name=dify_ai" target="_blank">
         <img src="https://img.shields.io/twitter/follow/dify_ai?logo=X&color=%20%23f5f5f5"
-            alt="suivre sur Twitter"></a>
+            alt="suivre sur X(Twitter)"></a>
     <a href="https://hub.docker.com/u/langgenius" target="_blank">
         <img alt="Tirages Docker" src="https://img.shields.io/docker/pulls/langgenius/dify-web?labelColor=%20%23FDB062&color=%20%23f79009"></a>
     <a href="https://github.com/langgenius/dify/graphs/commit-activity" target="_blank">
@@ -202,10 +202,13 @@ Si vous souhaitez configurer une configuration haute disponibilité, la communau
 
 #### Utilisation de Terraform pour le déploiement
 
+Déployez Dify sur une plateforme cloud en un clic en utilisant [terraform](https://www.terraform.io/)
+
 ##### Azure Global
-Utilisez [terraform](https://www.terraform.io/) pour déployer Dify sur Azure en un clic.
 - [Azure Terraform par @nikawang](https://github.com/nikawang/dify-azure-terraform)
 
+##### Google Cloud
+- [Google Cloud Terraform par @sotazum](https://github.com/DeNA/dify-google-cloud-terraform)
 
 ## Contribuer
 
@@ -226,7 +229,7 @@ Dans le même temps, veuillez envisager de soutenir Dify en le partageant sur le
 * [Discussion GitHub](https://github.com/langgenius/dify/discussions). Meilleur pour: partager des commentaires et poser des questions.
 * [Problèmes GitHub](https://github.com/langgenius/dify/issues). Meilleur pour: les bogues que vous rencontrez en utilisant Dify.AI et les propositions de fonctionnalités. Consultez notre [Guide de contribution](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md).
 * [Discord](https://discord.gg/FngNHpbcY7). Meilleur pour: partager vos applications et passer du temps avec la communauté.
-* [Twitter](https://twitter.com/dify_ai). Meilleur pour: partager vos applications et passer du temps avec la communauté.
+* [X(Twitter)](https://twitter.com/dify_ai). Meilleur pour: partager vos applications et passer du temps avec la communauté.
 
 ## Historique des étoiles
 

README_JA.md

@@ -17,7 +17,7 @@
             alt="Discordでチャット"></a>
     <a href="https://twitter.com/intent/follow?screen_name=dify_ai" target="_blank">
         <img src="https://img.shields.io/twitter/follow/dify_ai?logo=X&color=%20%23f5f5f5"
-            alt="Twitterでフォロー"></a>
+            alt="X(Twitter)でフォロー"></a>
     <a href="https://hub.docker.com/u/langgenius" target="_blank">
         <img alt="Docker Pulls" src="https://img.shields.io/docker/pulls/langgenius/dify-web?labelColor=%20%23FDB062&color=%20%23f79009"></a>
     <a href="https://github.com/langgenius/dify/graphs/commit-activity" target="_blank">
@@ -201,10 +201,13 @@ docker compose up -d
 
 #### Terraformを使用したデプロイ
 
+[terraform](https://www.terraform.io/) を使用して、ワンクリックでDifyをクラウドプラットフォームにデプロイします
+
 ##### Azure Global
-[terraform](https://www.terraform.io/) を使用して、AzureにDifyをワンクリックでデプロイします。
+- [@nikawangによるAzure Terraform](https://github.com/nikawang/dify-azure-terraform)
-- [nikawangのAzure Terraform](https://github.com/nikawang/dify-azure-terraform)
 
+##### Google Cloud
+- [@sotazumによるGoogle Cloud Terraform](https://github.com/DeNA/dify-google-cloud-terraform)
 
 ## 貢献
 
@@ -225,7 +228,7 @@ docker compose up -d
 * [Github Discussion](https://github.com/langgenius/dify/discussions). 主に: フィードバックの共有や質問。
 * [GitHub Issues](https://github.com/langgenius/dify/issues). 主に: Dify.AIを使用する際に発生するエラーや問題については、[貢献ガイド](CONTRIBUTING_JA.md)を参照してください
 * [Discord](https://discord.gg/FngNHpbcY7). 主に: アプリケーションの共有やコミュニティとの交流。
-* [Twitter](https://twitter.com/dify_ai). 主に: アプリケーションの共有やコミュニティとの交流。
+* [X(Twitter)](https://twitter.com/dify_ai). 主に: アプリケーションの共有やコミュニティとの交流。
 
 
 

README_KL.md

@@ -17,7 +17,7 @@
             alt="chat on Discord"></a>
     <a href="https://twitter.com/intent/follow?screen_name=dify_ai" target="_blank">
         <img src="https://img.shields.io/twitter/follow/dify_ai?logo=X&color=%20%23f5f5f5"
-            alt="follow on Twitter"></a>
+            alt="follow on X(Twitter)"></a>
     <a href="https://hub.docker.com/u/langgenius" target="_blank">
         <img alt="Docker Pulls" src="https://img.shields.io/docker/pulls/langgenius/dify-web?labelColor=%20%23FDB062&color=%20%23f79009"></a>
     <a href="https://github.com/langgenius/dify/graphs/commit-activity" target="_blank">
@@ -202,10 +202,13 @@ If you'd like to configure a highly-available setup, there are community-contrib
 
 #### Terraform atorlugu pilersitsineq
 
+wa'logh nIqHom neH ghun deployment toy'wI' [terraform](https://www.terraform.io/) lo'laH.
+
 ##### Azure Global
-Atoruk [terraform](https://www.terraform.io/) Dify-mik Azure-mut ataatsikkut ikkussuilluarlugu.
+- [Azure Terraform mung @nikawang](https://github.com/nikawang/dify-azure-terraform)
-- [Azure Terraform atorlugu @nikawang](https://github.com/nikawang/dify-azure-terraform)
 
+##### Google Cloud
+- [Google Cloud Terraform qachlot @sotazum](https://github.com/DeNA/dify-google-cloud-terraform)
 
 ## Contributing
 
@@ -228,7 +231,7 @@ At the same time, please consider supporting Dify by sharing it on social media
 ). Best for: sharing feedback and asking questions.
 * [GitHub Issues](https://github.com/langgenius/dify/issues). Best for: bugs you encounter using Dify.AI, and feature proposals. See our [Contribution Guide](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md).
 * [Discord](https://discord.gg/FngNHpbcY7). Best for: sharing your applications and hanging out with the community.
-* [Twitter](https://twitter.com/dify_ai). Best for: sharing your applications and hanging out with the community.
+* [X(Twitter)](https://twitter.com/dify_ai). Best for: sharing your applications and hanging out with the community.
 
 ## Star History
 

README_KR.md

@@ -17,7 +17,7 @@
             alt="chat on Discord"></a>
     <a href="https://twitter.com/intent/follow?screen_name=dify_ai" target="_blank">
         <img src="https://img.shields.io/twitter/follow/dify_ai?logo=X&color=%20%23f5f5f5"
-            alt="follow on Twitter"></a>
+            alt="follow on X(Twitter)"></a>
     <a href="https://hub.docker.com/u/langgenius" target="_blank">
         <img alt="Docker Pulls" src="https://img.shields.io/docker/pulls/langgenius/dify-web?labelColor=%20%23FDB062&color=%20%23f79009"></a>
     <a href="https://github.com/langgenius/dify/graphs/commit-activity" target="_blank">
@@ -39,7 +39,6 @@
   <a href="./README_AR.md"><img alt="README بالعربية" src="https://img.shields.io/badge/العربية-d9d9d9"></a>
   <a href="./README_TR.md"><img alt="Türkçe README" src="https://img.shields.io/badge/Türkçe-d9d9d9"></a>
   <a href="./README_VI.md"><img alt="README Tiếng Việt" src="https://img.shields.io/badge/Ti%E1%BA%BFng%20Vi%E1%BB%87t-d9d9d9"></a>
-
 </p>
 
 
@@ -195,10 +194,14 @@ Dify를 Kubernetes에 배포하고 프리미엄 스케일링 설정을 구성했
 
 #### Terraform을 사용한 배포
 
+[terraform](https://www.terraform.io/)을 사용하여 단 한 번의 클릭으로 Dify를 클라우드 플랫폼에 배포하십시오
+
 ##### Azure Global
-[terraform](https://www.terraform.io/)을 사용하여 Azure에 Dify를 원클릭으로 배포하세요.
 - [nikawang의 Azure Terraform](https://github.com/nikawang/dify-azure-terraform)
 
+##### Google Cloud
+- [sotazum의 Google Cloud Terraform](https://github.com/DeNA/dify-google-cloud-terraform)
+
 ## 기여
 
 코드에 기여하고 싶은 분들은 [기여 가이드](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md)를 참조하세요.

README_TR.md

@@ -17,7 +17,7 @@
             alt="Discord'da sohbet et"></a>
     <a href="https://twitter.com/intent/follow?screen_name=dify_ai" target="_blank">
         <img src="https://img.shields.io/twitter/follow/dify_ai?logo=X&color=%20%23f5f5f5"
-            alt="Twitter'da takip et"></a>
+            alt="X(Twitter)'da takip et"></a>
     <a href="https://hub.docker.com/u/langgenius" target="_blank">
         <img alt="Docker Çekmeleri" src="https://img.shields.io/docker/pulls/langgenius/dify-web?labelColor=%20%23FDB062&color=%20%23f79009"></a>
     <a href="https://github.com/langgenius/dify/graphs/commit-activity" target="_blank">
@@ -200,9 +200,13 @@ Yüksek kullanılabilirliğe sahip bir kurulum yapılandırmak isterseniz, Dify'
 
 #### Dağıtım için Terraform Kullanımı
 
+Dify'ı bulut platformuna tek tıklamayla dağıtın [terraform](https://www.terraform.io/) kullanarak
+
 ##### Azure Global
-[Terraform](https://www.terraform.io/) kullanarak Dify'ı Azure'a tek tıklamayla dağıtın.
+- [Azure Terraform tarafından @nikawang](https://github.com/nikawang/dify-azure-terraform)
-- [@nikawang tarafından Azure Terraform](https://github.com/nikawang/dify-azure-terraform)
+
+##### Google Cloud
+- [Google Cloud Terraform tarafından @sotazum](https://github.com/DeNA/dify-google-cloud-terraform)
 
 ## Katkıda Bulunma
 
@@ -222,7 +226,7 @@ Aynı zamanda, lütfen Dify'ı sosyal medyada, etkinliklerde ve konferanslarda p
 * [Github Tartışmaları](https://github.com/langgenius/dify/discussions). En uygun: geri bildirim paylaşmak ve soru sormak için.
 * [GitHub Sorunları](https://github.com/langgenius/dify/issues). En uygun: Dify.AI kullanırken karşılaştığınız hatalar ve özellik önerileri için. [Katkı Kılavuzumuza](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md) bakın.
 * [Discord](https://discord.gg/FngNHpbcY7). En uygun: uygulamalarınızı paylaşmak ve toplulukla vakit geçirmek için.
-* [Twitter](https://twitter.com/dify_ai). En uygun: uygulamalarınızı paylaşmak ve toplulukla vakit geçirmek için.
+* [X(Twitter)](https://twitter.com/dify_ai). En uygun: uygulamalarınızı paylaşmak ve toplulukla vakit geçirmek için.
 
 ## Star history
 

README_VI.md

@@ -17,7 +17,7 @@
             alt="chat trên Discord"></a>
     <a href="https://twitter.com/intent/follow?screen_name=dify_ai" target="_blank">
         <img src="https://img.shields.io/twitter/follow/dify_ai?logo=X&color=%20%23f5f5f5"
-            alt="theo dõi trên Twitter"></a>
+            alt="theo dõi trên X(Twitter)"></a>
     <a href="https://hub.docker.com/u/langgenius" target="_blank">
         <img alt="Docker Pulls" src="https://img.shields.io/docker/pulls/langgenius/dify-web?labelColor=%20%23FDB062&color=%20%23f79009"></a>
     <a href="https://github.com/langgenius/dify/graphs/commit-activity" target="_blank">
@@ -196,10 +196,14 @@ Nếu bạn muốn cấu hình một cài đặt có độ sẵn sàng cao, có
 
 #### Sử dụng Terraform để Triển khai
 
+Triển khai Dify lên nền tảng đám mây với một cú nhấp chuột bằng cách sử dụng [terraform](https://www.terraform.io/)
+
 ##### Azure Global
-Triển khai Dify lên Azure chỉ với một cú nhấp chuột bằng cách sử dụng [terraform](https://www.terraform.io/).
 - [Azure Terraform bởi @nikawang](https://github.com/nikawang/dify-azure-terraform)
 
+##### Google Cloud
+- [Google Cloud Terraform bởi @sotazum](https://github.com/DeNA/dify-google-cloud-terraform)
+
 ## Đóng góp
 
 Đối với những người muốn đóng góp mã, xem [Hướng dẫn Đóng góp](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md) của chúng tôi. 
@@ -219,7 +223,7 @@ Triển khai Dify lên Azure chỉ với một cú nhấp chuột bằng cách s
 * [Thảo luận GitHub](https://github.com/langgenius/dify/discussions). Tốt nhất cho: chia sẻ phản hồi và đặt câu hỏi.
 * [Vấn đề GitHub](https://github.com/langgenius/dify/issues). Tốt nhất cho: lỗi bạn gặp phải khi sử dụng Dify.AI và đề xuất tính năng. Xem [Hướng dẫn Đóng góp](https://github.com/langgenius/dify/blob/main/CONTRIBUTING.md) của chúng tôi.
 * [Discord](https://discord.gg/FngNHpbcY7). Tốt nhất cho: chia sẻ ứng dụng của bạn và giao lưu với cộng đồng.
-* [Twitter](https://twitter.com/dify_ai). Tốt nhất cho: chia sẻ ứng dụng của bạn và giao lưu với cộng đồng.
+* [X(Twitter)](https://twitter.com/dify_ai). Tốt nhất cho: chia sẻ ứng dụng của bạn và giao lưu với cộng đồng.
 
 ## Lịch sử Yêu thích
 

api/.env.example

@@ -271,6 +271,9 @@ HTTP_REQUEST_MAX_WRITE_TIMEOUT=600
 HTTP_REQUEST_NODE_MAX_BINARY_SIZE=10485760
 HTTP_REQUEST_NODE_MAX_TEXT_SIZE=1048576
 
+# Respect X-* headers to redirect clients
+RESPECT_XFORWARD_HEADERS_ENABLED=false
+
 # Log file path
 LOG_FILE=
 

api/app.py

@@ -26,7 +26,7 @@ from commands import register_commands
 from configs import dify_config
 
 # DO NOT REMOVE BELOW
-from events import event_handlers
+from events import event_handlers  # noqa: F401
 from extensions import (
     ext_celery,
     ext_code_based_extension,
@@ -36,6 +36,7 @@ from extensions import (
     ext_login,
     ext_mail,
     ext_migrate,
+    ext_proxy_fix,
     ext_redis,
     ext_sentry,
     ext_storage,
@@ -45,7 +46,7 @@ from extensions.ext_login import login_manager
 from libs.passport import PassportService
 
 # TODO: Find a way to avoid importing models here
-from models import account, dataset, model, source, task, tool, tools, web
+from models import account, dataset, model, source, task, tool, tools, web  # noqa: F401
 from services.account_service import AccountService
 
 # DO NOT REMOVE ABOVE
@@ -156,6 +157,7 @@ def initialize_extensions(app):
     ext_mail.init_app(app)
     ext_hosting_provider.init_app(app)
     ext_sentry.init_app(app)
+    ext_proxy_fix.init_app(app)
 
 
 # Flask-Login configuration
@@ -181,10 +183,10 @@ def load_user_from_request(request_from_flask_login):
     decoded = PassportService().verify(auth_token)
     user_id = decoded.get("user_id")
 
-    account = AccountService.load_logged_in_account(account_id=user_id, token=auth_token)
+    logged_in_account = AccountService.load_logged_in_account(account_id=user_id, token=auth_token)
-    if account:
+    if logged_in_account:
-        contexts.tenant_id.set(account.current_tenant_id)
+        contexts.tenant_id.set(logged_in_account.current_tenant_id)
-    return account
+    return logged_in_account
 
 
 @login_manager.unauthorized_handler

api/configs/feature/__init__.py

@@ -247,6 +247,12 @@ class HttpConfig(BaseSettings):
         default=None,
     )
 
+    RESPECT_XFORWARD_HEADERS_ENABLED: bool = Field(
+        description="Enable or disable the X-Forwarded-For Proxy Fix middleware from Werkzeug"
+        " to respect X-* headers to redirect clients",
+        default=False,
+    )
+
 
 class InnerAPIConfig(BaseSettings):
     """

api/configs/middleware/__init__.py

@@ -191,6 +191,22 @@ class CeleryConfig(DatabaseConfig):
         return self.CELERY_BROKER_URL.startswith("rediss://") if self.CELERY_BROKER_URL else False
 
 
+class InternalTestConfig(BaseSettings):
+    """
+    Configuration settings for Internal Test
+    """
+
+    AWS_SECRET_ACCESS_KEY: Optional[str] = Field(
+        description="Internal test AWS secret access key",
+        default=None,
+    )
+
+    AWS_ACCESS_KEY_ID: Optional[str] = Field(
+        description="Internal test AWS access key ID",
+        default=None,
+    )
+
+
 class MiddlewareConfig(
     # place the configs in alphabet order
     CeleryConfig,
@@ -224,5 +240,6 @@ class MiddlewareConfig(
     TiDBVectorConfig,
     WeaviateConfig,
     ElasticsearchConfig,
+    InternalTestConfig,
 ):
     pass

api/controllers/console/app/conversation.py

@@ -188,6 +188,7 @@ class ChatConversationApi(Resource):
                         subquery.c.from_end_user_session_id.ilike(keyword_filter),
                     ),
                 )
+                .group_by(Conversation.id)
             )
 
         account = current_user

api/controllers/console/datasets/external.py

@@ -13,6 +13,7 @@ from libs.login import login_required
 from services.dataset_service import DatasetService
 from services.external_knowledge_service import ExternalDatasetService
 from services.hit_testing_service import HitTestingService
+from services.knowledge_service import ExternalDatasetTestService
 
 
 def _validate_name(name):
@@ -232,8 +233,31 @@ class ExternalKnowledgeHitTestingApi(Resource):
             raise InternalServerError(str(e))
 
 
+class BedrockRetrievalApi(Resource):
+    # this api is only for internal testing
+    def post(self):
+        parser = reqparse.RequestParser()
+        parser.add_argument("retrieval_setting", nullable=False, required=True, type=dict, location="json")
+        parser.add_argument(
+            "query",
+            nullable=False,
+            required=True,
+            type=str,
+        )
+        parser.add_argument("knowledge_id", nullable=False, required=True, type=str)
+        args = parser.parse_args()
+
+        # Call the knowledge retrieval service
+        result = ExternalDatasetTestService.knowledge_retrieval(
+            args["retrieval_setting"], args["query"], args["knowledge_id"]
+        )
+        return result, 200
+
+
 api.add_resource(ExternalKnowledgeHitTestingApi, "/datasets/<uuid:dataset_id>/external-hit-testing")
 api.add_resource(ExternalDatasetCreateApi, "/datasets/external")
 api.add_resource(ExternalApiTemplateListApi, "/datasets/external-knowledge-api")
 api.add_resource(ExternalApiTemplateApi, "/datasets/external-knowledge-api/<uuid:external_knowledge_api_id>")
 api.add_resource(ExternalApiUseCheckApi, "/datasets/external-knowledge-api/<uuid:external_knowledge_api_id>/use-check")
+# this api is only for internal test
+api.add_resource(BedrockRetrievalApi, "/test/retrieval")

api/controllers/console/workspace/model_providers.py

@@ -126,13 +126,12 @@ class ModelProviderIconApi(Resource):
     Get model provider icon
     """
 
-    @setup_required
-    @login_required
-    @account_initialization_required
     def get(self, provider: str, icon_type: str, lang: str):
         model_provider_service = ModelProviderService()
         icon, mimetype = model_provider_service.get_model_provider_icon(
-            provider=provider, icon_type=icon_type, lang=lang
+            provider=provider,
+            icon_type=icon_type,
+            lang=lang,
         )
 
         return send_file(io.BytesIO(icon), mimetype=mimetype)

api/core/agent/cot_agent_runner.py

@@ -369,7 +369,7 @@ class CotAgentRunner(BaseAgentRunner, ABC):
         return message
 
     def _organize_historic_prompt_messages(
-        self, current_session_messages: list[PromptMessage] = None
+        self, current_session_messages: Optional[list[PromptMessage]] = None
     ) -> list[PromptMessage]:
         """
         organize historic prompt messages

api/core/agent/cot_chat_agent_runner.py

@@ -27,7 +27,7 @@ class CotChatAgentRunner(CotAgentRunner):
 
         return SystemPromptMessage(content=system_prompt)
 
-    def _organize_user_query(self, query, prompt_messages: list[PromptMessage] = None) -> list[PromptMessage]:
+    def _organize_user_query(self, query, prompt_messages: list[PromptMessage]) -> list[PromptMessage]:
         """
         Organize user query
         """

api/core/agent/cot_completion_agent_runner.py

@@ -1,4 +1,5 @@
 import json
+from typing import Optional
 
 from core.agent.cot_agent_runner import CotAgentRunner
 from core.model_runtime.entities.message_entities import AssistantPromptMessage, PromptMessage, UserPromptMessage
@@ -21,7 +22,7 @@ class CotCompletionAgentRunner(CotAgentRunner):
 
         return system_prompt
 
-    def _organize_historic_prompt(self, current_session_messages: list[PromptMessage] = None) -> str:
+    def _organize_historic_prompt(self, current_session_messages: Optional[list[PromptMessage]] = None) -> str:
         """
         Organize historic prompt
         """

api/core/agent/fc_agent_runner.py

@@ -2,7 +2,7 @@ import json
 import logging
 from collections.abc import Generator
 from copy import deepcopy
-from typing import Any, Union
+from typing import Any, Optional, Union
 
 from core.agent.base_agent_runner import BaseAgentRunner
 from core.app.apps.base_app_queue_manager import PublishFrom
@@ -370,7 +370,7 @@ class FunctionCallAgentRunner(BaseAgentRunner):
         return tool_calls
 
     def _init_system_message(
-        self, prompt_template: str, prompt_messages: list[PromptMessage] = None
+        self, prompt_template: str, prompt_messages: Optional[list[PromptMessage]] = None
     ) -> list[PromptMessage]:
         """
         Initialize system message
@@ -385,7 +385,7 @@ class FunctionCallAgentRunner(BaseAgentRunner):
 
         return prompt_messages
 
-    def _organize_user_query(self, query, prompt_messages: list[PromptMessage] = None) -> list[PromptMessage]:
+    def _organize_user_query(self, query, prompt_messages: list[PromptMessage]) -> list[PromptMessage]:
         """
         Organize user query
         """

api/core/app/apps/advanced_chat/app_generator.py

@@ -113,6 +113,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
             # always enable retriever resource in debugger mode
             app_config.additional_features.show_retrieve_source = True
 
+        workflow_run_id = str(uuid.uuid4())
         # init application generate entity
         application_generate_entity = AdvancedChatAppGenerateEntity(
             task_id=str(uuid.uuid4()),
@@ -127,6 +128,7 @@ class AdvancedChatAppGenerator(MessageBasedAppGenerator):
             invoke_from=invoke_from,
             extras=extras,
             trace_manager=trace_manager,
+            workflow_run_id=workflow_run_id,
         )
         contexts.tenant_id.set(application_generate_entity.app_config.tenant_id)
 

api/core/app/apps/advanced_chat/app_runner.py

@@ -149,6 +149,9 @@ class AdvancedChatAppRunner(WorkflowBasedAppRunner):
                 SystemVariableKey.CONVERSATION_ID: self.conversation.id,
                 SystemVariableKey.USER_ID: user_id,
                 SystemVariableKey.DIALOGUE_COUNT: conversation_dialogue_count,
+                SystemVariableKey.APP_ID: app_config.app_id,
+                SystemVariableKey.WORKFLOW_ID: app_config.workflow_id,
+                SystemVariableKey.WORKFLOW_RUN_ID: self.application_generate_entity.workflow_run_id,
             }
 
             # init variable pool

api/core/app/apps/advanced_chat/generate_task_pipeline.py

@@ -45,6 +45,7 @@ from core.app.entities.task_entities import (
 from core.app.task_pipeline.based_generate_task_pipeline import BasedGenerateTaskPipeline
 from core.app.task_pipeline.message_cycle_manage import MessageCycleManage
 from core.app.task_pipeline.workflow_cycle_manage import WorkflowCycleManage
+from core.model_runtime.entities.llm_entities import LLMUsage
 from core.model_runtime.utils.encoders import jsonable_encoder
 from core.ops.ops_trace_manager import TraceQueueManager
 from core.workflow.enums import SystemVariableKey
@@ -107,6 +108,10 @@ class AdvancedChatAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCyc
             SystemVariableKey.FILES: application_generate_entity.files,
             SystemVariableKey.CONVERSATION_ID: conversation.id,
             SystemVariableKey.USER_ID: user_id,
+            SystemVariableKey.DIALOGUE_COUNT: conversation.dialogue_count,
+            SystemVariableKey.APP_ID: application_generate_entity.app_config.app_id,
+            SystemVariableKey.WORKFLOW_ID: workflow.id,
+            SystemVariableKey.WORKFLOW_RUN_ID: application_generate_entity.workflow_run_id,
         }
 
         self._task_state = WorkflowTaskState()
@@ -505,6 +510,10 @@ class AdvancedChatAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCyc
             self._message.total_price = usage.total_price
             self._message.currency = usage.currency
 
+            self._task_state.metadata["usage"] = jsonable_encoder(usage)
+        else:
+            self._task_state.metadata["usage"] = jsonable_encoder(LLMUsage.empty_usage())
+
         db.session.commit()
 
         message_was_created.send(

api/core/app/apps/workflow/app_generator.py

@@ -99,6 +99,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
         user_id = user.id if isinstance(user, Account) else user.session_id
         trace_manager = TraceQueueManager(app_model.id, user_id)
 
+        workflow_run_id = str(uuid.uuid4())
         # init application generate entity
         application_generate_entity = WorkflowAppGenerateEntity(
             task_id=str(uuid.uuid4()),
@@ -110,6 +111,7 @@ class WorkflowAppGenerator(BaseAppGenerator):
             invoke_from=invoke_from,
             call_depth=call_depth,
             trace_manager=trace_manager,
+            workflow_run_id=workflow_run_id,
         )
         contexts.tenant_id.set(application_generate_entity.app_config.tenant_id)
 

api/core/app/apps/workflow/app_runner.py

@@ -90,6 +90,9 @@ class WorkflowAppRunner(WorkflowBasedAppRunner):
             system_inputs = {
                 SystemVariableKey.FILES: files,
                 SystemVariableKey.USER_ID: user_id,
+                SystemVariableKey.APP_ID: app_config.app_id,
+                SystemVariableKey.WORKFLOW_ID: app_config.workflow_id,
+                SystemVariableKey.WORKFLOW_RUN_ID: self.application_generate_entity.workflow_run_id,
             }
 
             variable_pool = VariablePool(

api/core/app/apps/workflow/generate_task_pipeline.py

@@ -97,6 +97,9 @@ class WorkflowAppGenerateTaskPipeline(BasedGenerateTaskPipeline, WorkflowCycleMa
         self._workflow_system_variables = {
             SystemVariableKey.FILES: application_generate_entity.files,
             SystemVariableKey.USER_ID: user_id,
+            SystemVariableKey.APP_ID: application_generate_entity.app_config.app_id,
+            SystemVariableKey.WORKFLOW_ID: workflow.id,
+            SystemVariableKey.WORKFLOW_RUN_ID: application_generate_entity.workflow_run_id,
         }
 
         self._task_state = WorkflowTaskState()

api/core/app/entities/app_invoke_entities.py

@@ -152,6 +152,7 @@ class AdvancedChatAppGenerateEntity(AppGenerateEntity):
 
     conversation_id: Optional[str] = None
     parent_message_id: Optional[str] = None
+    workflow_run_id: Optional[str] = None
     query: str
 
     class SingleIterationRunEntity(BaseModel):
@@ -172,6 +173,7 @@ class WorkflowAppGenerateEntity(AppGenerateEntity):
 
     # app config
     app_config: WorkflowUIBasedAppConfig
+    workflow_run_id: Optional[str] = None
 
     class SingleIterationRunEntity(BaseModel):
         """

api/core/app/task_pipeline/message_cycle_manage.py

@@ -82,8 +82,8 @@ class MessageCycleManage:
                 try:
                     name = LLMGenerator.generate_conversation_name(app_model.tenant_id, query)
                     conversation.name = name
-                except:
+                except Exception as e:
-                    pass
+                    logging.exception(f"generate conversation name failed: {e}")
 
                 db.session.merge(conversation)
                 db.session.commit()

api/core/app/task_pipeline/workflow_cycle_manage.py

@@ -85,6 +85,9 @@ class WorkflowCycleManage:
 
         # init workflow run
         workflow_run = WorkflowRun()
+        workflow_run_id = self._workflow_system_variables[SystemVariableKey.WORKFLOW_RUN_ID]
+        if workflow_run_id:
+            workflow_run.id = workflow_run_id
         workflow_run.tenant_id = self._workflow.tenant_id
         workflow_run.app_id = self._workflow.app_id
         workflow_run.sequence_number = new_sequence_number

api/core/callback_handler/agent_tool_callback_handler.py

@@ -1,9 +1,9 @@
-import os
 from collections.abc import Mapping, Sequence
 from typing import Any, Optional, TextIO, Union
 
 from pydantic import BaseModel
 
+from configs import dify_config
 from core.ops.entities.trace_entity import TraceTaskName
 from core.ops.ops_trace_manager import TraceQueueManager, TraceTask
 from core.tools.entities.tool_entities import ToolInvokeMessage
@@ -50,7 +50,8 @@ class DifyAgentCallbackHandler(BaseModel):
         tool_inputs: Mapping[str, Any],
     ) -> None:
         """Do nothing."""
-        print_text("\n[on_tool_start] ToolCall:" + tool_name + "\n" + str(tool_inputs) + "\n", color=self.color)
+        if dify_config.DEBUG:
+            print_text("\n[on_tool_start] ToolCall:" + tool_name + "\n" + str(tool_inputs) + "\n", color=self.color)
 
     def on_tool_end(
         self,
@@ -62,11 +63,12 @@ class DifyAgentCallbackHandler(BaseModel):
         trace_manager: Optional[TraceQueueManager] = None,
     ) -> None:
         """If not the final action, print out observation."""
-        print_text("\n[on_tool_end]\n", color=self.color)
+        if dify_config.DEBUG:
-        print_text("Tool: " + tool_name + "\n", color=self.color)
+            print_text("\n[on_tool_end]\n", color=self.color)
-        print_text("Inputs: " + str(tool_inputs) + "\n", color=self.color)
+            print_text("Tool: " + tool_name + "\n", color=self.color)
-        print_text("Outputs: " + str(tool_outputs)[:1000] + "\n", color=self.color)
+            print_text("Inputs: " + str(tool_inputs) + "\n", color=self.color)
-        print_text("\n")
+            print_text("Outputs: " + str(tool_outputs)[:1000] + "\n", color=self.color)
+            print_text("\n")
 
         if trace_manager:
             trace_manager.add_trace_task(
@@ -82,30 +84,33 @@ class DifyAgentCallbackHandler(BaseModel):
 
     def on_tool_error(self, error: Union[Exception, KeyboardInterrupt], **kwargs: Any) -> None:
         """Do nothing."""
-        print_text("\n[on_tool_error] Error: " + str(error) + "\n", color="red")
+        if dify_config.DEBUG:
+            print_text("\n[on_tool_error] Error: " + str(error) + "\n", color="red")
 
     def on_agent_start(self, thought: str) -> None:
         """Run on agent start."""
-        if thought:
+        if dify_config.DEBUG:
-            print_text(
+            if thought:
-                "\n[on_agent_start] \nCurrent Loop: " + str(self.current_loop) + "\nThought: " + thought + "\n",
+                print_text(
-                color=self.color,
+                    "\n[on_agent_start] \nCurrent Loop: " + str(self.current_loop) + "\nThought: " + thought + "\n",
-            )
+                    color=self.color,
-        else:
+                )
-            print_text("\n[on_agent_start] \nCurrent Loop: " + str(self.current_loop) + "\n", color=self.color)
+            else:
+                print_text("\n[on_agent_start] \nCurrent Loop: " + str(self.current_loop) + "\n", color=self.color)
 
     def on_agent_finish(self, color: Optional[str] = None, **kwargs: Any) -> None:
         """Run on agent end."""
-        print_text("\n[on_agent_finish]\n Loop: " + str(self.current_loop) + "\n", color=self.color)
+        if dify_config.DEBUG:
+            print_text("\n[on_agent_finish]\n Loop: " + str(self.current_loop) + "\n", color=self.color)
 
         self.current_loop += 1
 
     @property
     def ignore_agent(self) -> bool:
         """Whether to ignore agent callbacks."""
-        return not os.environ.get("DEBUG") or os.environ.get("DEBUG").lower() != "true"
+        return not dify_config.DEBUG
 
     @property
     def ignore_chat_model(self) -> bool:
         """Whether to ignore chat model callbacks."""
-        return not os.environ.get("DEBUG") or os.environ.get("DEBUG").lower() != "true"
+        return not dify_config.DEBUG

api/core/file/message_file_parser.py

@@ -198,16 +198,34 @@ class MessageFileParser:
                     if "amazonaws.com" not in parsed_url.netloc:
                         return False
                     query_params = parse_qs(parsed_url.query)
-                    required_params = ["Signature", "Expires"]
+
-                    for param in required_params:
+                    def check_presign_v2(query_params):
-                        if param not in query_params:
+                        required_params = ["Signature", "Expires"]
+                        for param in required_params:
+                            if param not in query_params:
+                                return False
+                        if not query_params["Expires"][0].isdigit():
                             return False
-                    if not query_params["Expires"][0].isdigit():
+                        signature = query_params["Signature"][0]
-                        return False
+                        if not re.match(r"^[A-Za-z0-9+/]+={0,2}$", signature):
-                    signature = query_params["Signature"][0]
+                            return False
-                    if not re.match(r"^[A-Za-z0-9+/]+={0,2}$", signature):
+
-                        return False
+                        return True
-                    return True
+
+                    def check_presign_v4(query_params):
+                        required_params = ["X-Amz-Signature", "X-Amz-Expires"]
+                        for param in required_params:
+                            if param not in query_params:
+                                return False
+                        if not query_params["X-Amz-Expires"][0].isdigit():
+                            return False
+                        signature = query_params["X-Amz-Signature"][0]
+                        if not re.match(r"^[A-Za-z0-9+/]+={0,2}$", signature):
+                            return False
+
+                        return True
+
+                    return check_presign_v4(query_params) or check_presign_v2(query_params)
                 except Exception:
                     return False
 

api/core/indexing_runner.py

@@ -211,9 +211,9 @@ class IndexingRunner:
         tenant_id: str,
         extract_settings: list[ExtractSetting],
         tmp_processing_rule: dict,
-        doc_form: str = None,
+        doc_form: Optional[str] = None,
         doc_language: str = "English",
-        dataset_id: str = None,
+        dataset_id: Optional[str] = None,
         indexing_technique: str = "economy",
     ) -> dict:
         """

api/core/memory/token_buffer_memory.py

@@ -58,7 +58,11 @@ class TokenBufferMemory:
         # instead of all messages from the conversation, we only need to extract messages
         # that belong to the thread of last message
         thread_messages = extract_thread_messages(messages)
-        thread_messages.pop(0)
+
+        # for newly created message, its answer is temporarily empty, we don't need to add it to memory
+        if thread_messages and not thread_messages[-1].answer:
+            thread_messages.pop()
+
         messages = list(reversed(thread_messages))
 
         message_file_parser = MessageFileParser(tenant_id=app_record.tenant_id, app_id=app_record.id)

api/core/model_runtime/model_providers/__base/large_language_model.py

@@ -94,7 +94,7 @@ class LargeLanguageModel(AIModel):
         )
 
         try:
-            if "response_format" in model_parameters:
+            if "response_format" in model_parameters and model_parameters["response_format"] in {"JSON", "XML"}:
                 result = self._code_block_mode_wrapper(
                     model=model,
                     credentials=credentials,

api/core/model_runtime/model_providers/__base/tts_model.py

@@ -1,7 +1,7 @@
 import logging
 import re
 from abc import abstractmethod
-from typing import Optional
+from typing import Any, Optional
 
 from pydantic import ConfigDict
 
@@ -88,7 +88,7 @@ class TTSModel(AIModel):
             else:
                 return [{"name": d["name"], "value": d["mode"]} for d in voices]
 
-    def _get_model_default_voice(self, model: str, credentials: dict) -> any:
+    def _get_model_default_voice(self, model: str, credentials: dict) -> Any:
         """
         Get voice for given tts model
 

api/core/model_runtime/model_providers/anthropic/llm/llm.py

@@ -169,7 +169,7 @@ class AnthropicLargeLanguageModel(LargeLanguageModel):
         stop: Optional[list[str]] = None,
         stream: bool = True,
         user: Optional[str] = None,
-        callbacks: list[Callback] = None,
+        callbacks: Optional[list[Callback]] = None,
     ) -> Union[LLMResult, Generator]:
         """
         Code block mode wrapper for invoking large language model

api/core/model_runtime/model_providers/azure_openai/_constant.py

@@ -1081,8 +1081,81 @@ LLM_BASE_MODELS = [
             ),
         ),
     ),
+    AzureBaseModel(
+        base_model_name="o1-preview",
+        entity=AIModelEntity(
+            model="fake-deployment-name",
+            label=I18nObject(
+                en_US="fake-deployment-name-label",
+            ),
+            model_type=ModelType.LLM,
+            features=[
+                ModelFeature.AGENT_THOUGHT,
+            ],
+            fetch_from=FetchFrom.CUSTOMIZABLE_MODEL,
+            model_properties={
+                ModelPropertyKey.MODE: LLMMode.CHAT.value,
+                ModelPropertyKey.CONTEXT_SIZE: 128000,
+            },
+            parameter_rules=[
+                ParameterRule(
+                    name="response_format",
+                    label=I18nObject(zh_Hans="回复格式", en_US="response_format"),
+                    type="string",
+                    help=I18nObject(
+                        zh_Hans="指定模型必须输出的格式", en_US="specifying the format that the model must output"
+                    ),
+                    required=False,
+                    options=["text", "json_object"],
+                ),
+                _get_max_tokens(default=512, min_val=1, max_val=32768),
+            ],
+            pricing=PriceConfig(
+                input=15.00,
+                output=60.00,
+                unit=0.000001,
+                currency="USD",
+            ),
+        ),
+    ),
+    AzureBaseModel(
+        base_model_name="o1-mini",
+        entity=AIModelEntity(
+            model="fake-deployment-name",
+            label=I18nObject(
+                en_US="fake-deployment-name-label",
+            ),
+            model_type=ModelType.LLM,
+            features=[
+                ModelFeature.AGENT_THOUGHT,
+            ],
+            fetch_from=FetchFrom.CUSTOMIZABLE_MODEL,
+            model_properties={
+                ModelPropertyKey.MODE: LLMMode.CHAT.value,
+                ModelPropertyKey.CONTEXT_SIZE: 128000,
+            },
+            parameter_rules=[
+                ParameterRule(
+                    name="response_format",
+                    label=I18nObject(zh_Hans="回复格式", en_US="response_format"),
+                    type="string",
+                    help=I18nObject(
+                        zh_Hans="指定模型必须输出的格式", en_US="specifying the format that the model must output"
+                    ),
+                    required=False,
+                    options=["text", "json_object"],
+                ),
+                _get_max_tokens(default=512, min_val=1, max_val=65536),
+            ],
+            pricing=PriceConfig(
+                input=3.00,
+                output=12.00,
+                unit=0.000001,
+                currency="USD",
+            ),
+        ),
+    ),
 ]
-
 EMBEDDING_BASE_MODELS = [
     AzureBaseModel(
         base_model_name="text-embedding-ada-002",

api/core/model_runtime/model_providers/azure_openai/azure_openai.yaml

@@ -53,6 +53,9 @@ model_credential_schema:
       type: select
       required: true
       options:
+        - label:
+            en_US: 2024-09-01-preview
+          value: 2024-09-01-preview
         - label:
             en_US: 2024-08-01-preview
           value: 2024-08-01-preview
@@ -120,6 +123,18 @@ model_credential_schema:
           show_on:
             - variable: __model_type
               value: llm
+        - label:
+            en_US: o1-mini
+          value: o1-mini
+          show_on:
+            - variable: __model_type
+              value: llm
+        - label:
+            en_US: o1-preview
+          value: o1-preview
+          show_on:
+            - variable: __model_type
+              value: llm
         - label:
             en_US: gpt-4o-mini
           value: gpt-4o-mini

api/core/model_runtime/model_providers/azure_openai/llm/llm.py

@@ -312,10 +312,24 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
         if user:
             extra_model_kwargs["user"] = user
 
+        # clear illegal prompt messages
+        prompt_messages = self._clear_illegal_prompt_messages(model, prompt_messages)
+
+        block_as_stream = False
+        if model.startswith("o1"):
+            if stream:
+                block_as_stream = True
+                stream = False
+
+                if "stream_options" in extra_model_kwargs:
+                    del extra_model_kwargs["stream_options"]
+
+            if "stop" in extra_model_kwargs:
+                del extra_model_kwargs["stop"]
+
         # chat model
-        messages = [self._convert_prompt_message_to_dict(m) for m in prompt_messages]
         response = client.chat.completions.create(
-            messages=messages,
+            messages=[self._convert_prompt_message_to_dict(m) for m in prompt_messages],
             model=model,
             stream=stream,
             **model_parameters,
@@ -325,7 +339,91 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
         if stream:
             return self._handle_chat_generate_stream_response(model, credentials, response, prompt_messages, tools)
 
-        return self._handle_chat_generate_response(model, credentials, response, prompt_messages, tools)
+        block_result = self._handle_chat_generate_response(model, credentials, response, prompt_messages, tools)
+
+        if block_as_stream:
+            return self._handle_chat_block_as_stream_response(block_result, prompt_messages, stop)
+
+        return block_result
+
+    def _handle_chat_block_as_stream_response(
+        self,
+        block_result: LLMResult,
+        prompt_messages: list[PromptMessage],
+        stop: Optional[list[str]] = None,
+    ) -> Generator[LLMResultChunk, None, None]:
+        """
+        Handle llm chat response
+
+        :param model: model name
+        :param credentials: credentials
+        :param response: response
+        :param prompt_messages: prompt messages
+        :param tools: tools for tool calling
+        :param stop: stop words
+        :return: llm response chunk generator
+        """
+        text = block_result.message.content
+        text = cast(str, text)
+
+        if stop:
+            text = self.enforce_stop_tokens(text, stop)
+
+        yield LLMResultChunk(
+            model=block_result.model,
+            prompt_messages=prompt_messages,
+            system_fingerprint=block_result.system_fingerprint,
+            delta=LLMResultChunkDelta(
+                index=0,
+                message=AssistantPromptMessage(content=text),
+                finish_reason="stop",
+                usage=block_result.usage,
+            ),
+        )
+
+    def _clear_illegal_prompt_messages(self, model: str, prompt_messages: list[PromptMessage]) -> list[PromptMessage]:
+        """
+        Clear illegal prompt messages for OpenAI API
+
+        :param model: model name
+        :param prompt_messages: prompt messages
+        :return: cleaned prompt messages
+        """
+        checklist = ["gpt-4-turbo", "gpt-4-turbo-2024-04-09"]
+
+        if model in checklist:
+            # count how many user messages are there
+            user_message_count = len([m for m in prompt_messages if isinstance(m, UserPromptMessage)])
+            if user_message_count > 1:
+                for prompt_message in prompt_messages:
+                    if isinstance(prompt_message, UserPromptMessage):
+                        if isinstance(prompt_message.content, list):
+                            prompt_message.content = "\n".join(
+                                [
+                                    item.data
+                                    if item.type == PromptMessageContentType.TEXT
+                                    else "[IMAGE]"
+                                    if item.type == PromptMessageContentType.IMAGE
+                                    else ""
+                                    for item in prompt_message.content
+                                ]
+                            )
+
+        if model.startswith("o1"):
+            system_message_count = len([m for m in prompt_messages if isinstance(m, SystemPromptMessage)])
+            if system_message_count > 0:
+                new_prompt_messages = []
+                for prompt_message in prompt_messages:
+                    if isinstance(prompt_message, SystemPromptMessage):
+                        prompt_message = UserPromptMessage(
+                            content=prompt_message.content,
+                            name=prompt_message.name,
+                        )
+
+                    new_prompt_messages.append(prompt_message)
+                prompt_messages = new_prompt_messages
+
+        return prompt_messages
 
     def _handle_chat_generate_response(
         self,
@@ -560,7 +658,7 @@ class AzureOpenAILargeLanguageModel(_CommonAzureOpenAI, LargeLanguageModel):
             tokens_per_message = 4
             # if there's a name, the role is omitted
             tokens_per_name = -1
-        elif model.startswith("gpt-35-turbo") or model.startswith("gpt-4"):
+        elif model.startswith("gpt-35-turbo") or model.startswith("gpt-4") or model.startswith("o1"):
             tokens_per_message = 3
             tokens_per_name = 1
         else:

api/core/model_runtime/model_providers/azure_openai/tts/tts.py

@@ -1,6 +1,6 @@
 import concurrent.futures
 import copy
-from typing import Optional
+from typing import Any, Optional
 
 from openai import AzureOpenAI
 
@@ -19,7 +19,7 @@ class AzureOpenAIText2SpeechModel(_CommonAzureOpenAI, TTSModel):
 
     def _invoke(
         self, model: str, tenant_id: str, credentials: dict, content_text: str, voice: str, user: Optional[str] = None
-    ) -> any:
+    ) -> Any:
         """
         _invoke text2speech model
 
@@ -56,7 +56,7 @@ class AzureOpenAIText2SpeechModel(_CommonAzureOpenAI, TTSModel):
         except Exception as ex:
             raise CredentialsValidateFailedError(str(ex))
 
-    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> any:
+    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> Any:
         """
         _tts_invoke_streaming text2speech model
         :param model: model name

api/core/model_runtime/model_providers/bedrock/llm/llm.py

@@ -92,7 +92,7 @@ class BedrockLargeLanguageModel(LargeLanguageModel):
         stop: Optional[list[str]] = None,
         stream: bool = True,
         user: Optional[str] = None,
-        callbacks: list[Callback] = None,
+        callbacks: Optional[list[Callback]] = None,
     ) -> Union[LLMResult, Generator]:
         """
         Code block mode wrapper for invoking large language model

api/core/model_runtime/model_providers/fireworks/llm/llm.py

@@ -511,7 +511,7 @@ class FireworksLargeLanguageModel(_CommonFireworks, LargeLanguageModel):
         model: str,
         messages: list[PromptMessage],
         tools: Optional[list[PromptMessageTool]] = None,
-        credentials: dict = None,
+        credentials: Optional[dict] = None,
     ) -> int:
         """
         Approximate num tokens with GPT2 tokenizer.

api/core/model_runtime/model_providers/fishaudio/tts/tts.py

@@ -1,4 +1,4 @@
-from typing import Optional
+from typing import Any, Optional
 
 import httpx
 
@@ -46,7 +46,7 @@ class FishAudioText2SpeechModel(TTSModel):
         content_text: str,
         voice: str,
         user: Optional[str] = None,
-    ) -> any:
+    ) -> Any:
         """
         Invoke text2speech model
 
@@ -87,7 +87,7 @@ class FishAudioText2SpeechModel(TTSModel):
         except Exception as ex:
             raise CredentialsValidateFailedError(str(ex))
 
-    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> any:
+    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> Any:
         """
         Invoke streaming text2speech model
         :param model: model name
@@ -112,7 +112,7 @@ class FishAudioText2SpeechModel(TTSModel):
         except Exception as ex:
             raise InvokeBadRequestError(str(ex))
 
-    def _tts_invoke_streaming_sentence(self, credentials: dict, content_text: str, voice: Optional[str] = None) -> any:
+    def _tts_invoke_streaming_sentence(self, credentials: dict, content_text: str, voice: Optional[str] = None) -> Any:
         """
         Invoke streaming text2speech model
 

api/core/model_runtime/model_providers/groq/llm/_position.yaml

@@ -5,3 +5,4 @@
 - llama3-8b-8192
 - mixtral-8x7b-32768
 - llama2-70b-4096
+- llama-guard-3-8b

api/core/model_runtime/model_providers/groq/llm/llama-guard-3-8b.yaml

@@ -0,0 +1,25 @@
+model: llama-guard-3-8b
+label:
+  zh_Hans: Llama-Guard-3-8B
+  en_US: Llama-Guard-3-8B
+model_type: llm
+features:
+  - agent-thought
+model_properties:
+  mode: chat
+  context_size: 8192
+parameter_rules:
+  - name: temperature
+    use_template: temperature
+  - name: top_p
+    use_template: top_p
+  - name: max_tokens
+    use_template: max_tokens
+    default: 512
+    min: 1
+    max: 8192
+pricing:
+  input: '0.20'
+  output: '0.20'
+  unit: '0.000001'
+  currency: USD

api/core/model_runtime/model_providers/jina/rerank/rerank.py

@@ -61,11 +61,19 @@ class JinaRerankModel(RerankModel):
 
             rerank_documents = []
             for result in results["results"]:
+                index = result["index"]
+                if "document" in result:
+                    text = result["document"]["text"]
+                else:
+                    # llama.cpp rerank maynot return original documents
+                    text = docs[index]
+
                 rerank_document = RerankDocument(
-                    index=result["index"],
+                    index=index,
-                    text=result["document"]["text"],
+                    text=text,
                     score=result["relevance_score"],
                 )
+
                 if score_threshold is None or result["relevance_score"] >= score_threshold:
                     rerank_documents.append(rerank_document)
 

api/core/model_runtime/model_providers/localai/rerank/rerank.py

@@ -70,11 +70,19 @@ class LocalaiRerankModel(RerankModel):
 
             rerank_documents = []
             for result in results["results"]:
+                index = result["index"]
+                if "document" in result:
+                    text = result["document"]["text"]
+                else:
+                    # llama.cpp rerank maynot return original documents
+                    text = docs[index]
+
                 rerank_document = RerankDocument(
-                    index=result["index"],
+                    index=index,
-                    text=result["document"]["text"],
+                    text=text,
                     score=result["relevance_score"],
                 )
+
                 if score_threshold is None or result["relevance_score"] >= score_threshold:
                     rerank_documents.append(rerank_document)
 

api/core/model_runtime/model_providers/openai/llm/llm.py

@@ -111,7 +111,7 @@ class OpenAILargeLanguageModel(_CommonOpenAI, LargeLanguageModel):
         stop: Optional[list[str]] = None,
         stream: bool = True,
         user: Optional[str] = None,
-        callbacks: list[Callback] = None,
+        callbacks: Optional[list[Callback]] = None,
     ) -> Union[LLMResult, Generator]:
         """
         Code block mode wrapper for invoking large language model

api/core/model_runtime/model_providers/openai/tts/tts.py

@@ -1,5 +1,5 @@
 import concurrent.futures
-from typing import Optional
+from typing import Any, Optional
 
 from openai import OpenAI
 
@@ -16,7 +16,7 @@ class OpenAIText2SpeechModel(_CommonOpenAI, TTSModel):
 
     def _invoke(
         self, model: str, tenant_id: str, credentials: dict, content_text: str, voice: str, user: Optional[str] = None
-    ) -> any:
+    ) -> Any:
         """
         _invoke text2speech model
 
@@ -55,7 +55,7 @@ class OpenAIText2SpeechModel(_CommonOpenAI, TTSModel):
         except Exception as ex:
             raise CredentialsValidateFailedError(str(ex))
 
-    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> any:
+    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> Any:
         """
         _tts_invoke_streaming text2speech model
 

api/core/model_runtime/model_providers/openai_api_compatible/llm/llm.py

@@ -688,7 +688,7 @@ class OAIAPICompatLargeLanguageModel(_CommonOaiApiCompat, LargeLanguageModel):
         model: str,
         messages: list[PromptMessage],
         tools: Optional[list[PromptMessageTool]] = None,
-        credentials: dict = None,
+        credentials: Optional[dict] = None,
     ) -> int:
         """
         Approximate num tokens with GPT2 tokenizer.

api/core/model_runtime/model_providers/openrouter/llm/claude-3-5-sonnet.yaml

@@ -27,9 +27,9 @@ parameter_rules:
   - name: max_tokens
     use_template: max_tokens
     required: true
-    default: 4096
+    default: 8192
     min: 1
-    max: 4096
+    max: 8192
   - name: response_format
     use_template: response_format
 pricing:

api/core/model_runtime/model_providers/sagemaker/tts/tts.py

@@ -77,7 +77,7 @@ class SageMakerText2SpeechModel(TTSModel):
         """
         pass
 
-    def _detect_lang_code(self, content: str, map_dict: dict = None):
+    def _detect_lang_code(self, content: str, map_dict: Optional[dict] = None):
         map_dict = {"zh": "<|zh|>", "en": "<|en|>", "ja": "<|jp|>", "zh-TW": "<|yue|>", "ko": "<|ko|>"}
 
         response = self.comprehend_client.detect_dominant_language(Text=content)
@@ -192,7 +192,7 @@ class SageMakerText2SpeechModel(TTSModel):
             InvokeBadRequestError: [InvokeBadRequestError, KeyError, ValueError],
         }
 
-    def _get_model_default_voice(self, model: str, credentials: dict) -> any:
+    def _get_model_default_voice(self, model: str, credentials: dict) -> Any:
         return ""
 
     def _get_model_word_limit(self, model: str, credentials: dict) -> int:
@@ -225,7 +225,7 @@ class SageMakerText2SpeechModel(TTSModel):
         json_obj = json.loads(json_str)
         return json_obj
 
-    def _tts_invoke_streaming(self, model_type: str, payload: dict, sagemaker_endpoint: str) -> any:
+    def _tts_invoke_streaming(self, model_type: str, payload: dict, sagemaker_endpoint: str) -> Any:
         """
         _tts_invoke_streaming text2speech model
 

api/core/model_runtime/model_providers/siliconflow/llm/llm.py

@@ -1,8 +1,18 @@
 from collections.abc import Generator
 from typing import Optional, Union
 
-from core.model_runtime.entities.llm_entities import LLMResult
+from core.model_runtime.entities.common_entities import I18nObject
+from core.model_runtime.entities.llm_entities import LLMMode, LLMResult
 from core.model_runtime.entities.message_entities import PromptMessage, PromptMessageTool
+from core.model_runtime.entities.model_entities import (
+    AIModelEntity,
+    FetchFrom,
+    ModelFeature,
+    ModelPropertyKey,
+    ModelType,
+    ParameterRule,
+    ParameterType,
+)
 from core.model_runtime.model_providers.openai_api_compatible.llm.llm import OAIAPICompatLargeLanguageModel
 
 
@@ -29,3 +39,53 @@ class SiliconflowLargeLanguageModel(OAIAPICompatLargeLanguageModel):
     def _add_custom_parameters(cls, credentials: dict) -> None:
         credentials["mode"] = "chat"
         credentials["endpoint_url"] = "https://api.siliconflow.cn/v1"
+
+    def get_customizable_model_schema(self, model: str, credentials: dict) -> AIModelEntity | None:
+        return AIModelEntity(
+            model=model,
+            label=I18nObject(en_US=model, zh_Hans=model),
+            model_type=ModelType.LLM,
+            features=[ModelFeature.TOOL_CALL, ModelFeature.MULTI_TOOL_CALL, ModelFeature.STREAM_TOOL_CALL]
+            if credentials.get("function_calling_type") == "tool_call"
+            else [],
+            fetch_from=FetchFrom.CUSTOMIZABLE_MODEL,
+            model_properties={
+                ModelPropertyKey.CONTEXT_SIZE: int(credentials.get("context_size", 8000)),
+                ModelPropertyKey.MODE: LLMMode.CHAT.value,
+            },
+            parameter_rules=[
+                ParameterRule(
+                    name="temperature",
+                    use_template="temperature",
+                    label=I18nObject(en_US="Temperature", zh_Hans="温度"),
+                    type=ParameterType.FLOAT,
+                ),
+                ParameterRule(
+                    name="max_tokens",
+                    use_template="max_tokens",
+                    default=512,
+                    min=1,
+                    max=int(credentials.get("max_tokens", 1024)),
+                    label=I18nObject(en_US="Max Tokens", zh_Hans="最大标记"),
+                    type=ParameterType.INT,
+                ),
+                ParameterRule(
+                    name="top_p",
+                    use_template="top_p",
+                    label=I18nObject(en_US="Top P", zh_Hans="Top P"),
+                    type=ParameterType.FLOAT,
+                ),
+                ParameterRule(
+                    name="top_k",
+                    use_template="top_k",
+                    label=I18nObject(en_US="Top K", zh_Hans="Top K"),
+                    type=ParameterType.FLOAT,
+                ),
+                ParameterRule(
+                    name="frequency_penalty",
+                    use_template="frequency_penalty",
+                    label=I18nObject(en_US="Frequency Penalty", zh_Hans="重复惩罚"),
+                    type=ParameterType.FLOAT,
+                ),
+            ],
+        )

api/core/model_runtime/model_providers/siliconflow/siliconflow.yaml

@@ -20,6 +20,7 @@ supported_model_types:
   - speech2text
 configurate_methods:
   - predefined-model
+  - customizable-model
 provider_credential_schema:
   credential_form_schemas:
     - variable: api_key
@@ -30,3 +31,57 @@ provider_credential_schema:
       placeholder:
         zh_Hans: 在此输入您的 API Key
         en_US: Enter your API Key
+model_credential_schema:
+  model:
+    label:
+      en_US: Model Name
+      zh_Hans: 模型名称
+    placeholder:
+      en_US: Enter your model name
+      zh_Hans: 输入模型名称
+  credential_form_schemas:
+    - variable: api_key
+      label:
+        en_US: API Key
+      type: secret-input
+      required: true
+      placeholder:
+        zh_Hans: 在此输入您的 API Key
+        en_US: Enter your API Key
+    - variable: context_size
+      label:
+        zh_Hans: 模型上下文长度
+        en_US: Model context size
+      required: true
+      type: text-input
+      default: '4096'
+      placeholder:
+        zh_Hans: 在此输入您的模型上下文长度
+        en_US: Enter your Model context size
+    - variable: max_tokens
+      label:
+        zh_Hans: 最大 token 上限
+        en_US: Upper bound for max tokens
+      default: '4096'
+      type: text-input
+      show_on:
+        - variable: __model_type
+          value: llm
+    - variable: function_calling_type
+      label:
+        en_US: Function calling
+      type: select
+      required: false
+      default: no_call
+      options:
+        - value: no_call
+          label:
+            en_US: Not Support
+            zh_Hans: 不支持
+        - value: function_call
+          label:
+            en_US: Support
+            zh_Hans: 支持
+      show_on:
+        - variable: __model_type
+          value: llm

api/core/model_runtime/model_providers/tongyi/rerank/_position.yaml

@@ -0,0 +1 @@
+- gte-rerank

api/core/model_runtime/model_providers/tongyi/rerank/gte-rerank.yaml

@@ -0,0 +1,4 @@
+model: gte-rerank
+model_type: rerank
+model_properties:
+  context_size: 4000

api/core/model_runtime/model_providers/tongyi/rerank/rerank.py

@@ -0,0 +1,136 @@
+from typing import Optional
+
+import dashscope
+from dashscope.common.error import (
+    AuthenticationError,
+    InvalidParameter,
+    RequestFailure,
+    ServiceUnavailableError,
+    UnsupportedHTTPMethod,
+    UnsupportedModel,
+)
+
+from core.model_runtime.entities.rerank_entities import RerankDocument, RerankResult
+from core.model_runtime.errors.invoke import (
+    InvokeAuthorizationError,
+    InvokeBadRequestError,
+    InvokeConnectionError,
+    InvokeError,
+    InvokeRateLimitError,
+    InvokeServerUnavailableError,
+)
+from core.model_runtime.errors.validate import CredentialsValidateFailedError
+from core.model_runtime.model_providers.__base.rerank_model import RerankModel
+
+
+class GTERerankModel(RerankModel):
+    """
+    Model class for GTE rerank model.
+    """
+
+    def _invoke(
+        self,
+        model: str,
+        credentials: dict,
+        query: str,
+        docs: list[str],
+        score_threshold: Optional[float] = None,
+        top_n: Optional[int] = None,
+        user: Optional[str] = None,
+    ) -> RerankResult:
+        """
+        Invoke rerank model
+
+        :param model: model name
+        :param credentials: model credentials
+        :param query: search query
+        :param docs: docs for reranking
+        :param score_threshold: score threshold
+        :param top_n: top n
+        :param user: unique user id
+        :return: rerank result
+        """
+        if len(docs) == 0:
+            return RerankResult(model=model, docs=docs)
+
+        # initialize client
+        dashscope.api_key = credentials["dashscope_api_key"]
+
+        response = dashscope.TextReRank.call(
+            query=query,
+            documents=docs,
+            model=model,
+            top_n=top_n,
+            return_documents=True,
+        )
+
+        rerank_documents = []
+        for _, result in enumerate(response.output.results):
+            # format document
+            rerank_document = RerankDocument(
+                index=result.index,
+                score=result.relevance_score,
+                text=result["document"]["text"],
+            )
+
+            # score threshold check
+            if score_threshold is not None:
+                if result.relevance_score >= score_threshold:
+                    rerank_documents.append(rerank_document)
+            else:
+                rerank_documents.append(rerank_document)
+
+        return RerankResult(model=model, docs=rerank_documents)
+
+    def validate_credentials(self, model: str, credentials: dict) -> None:
+        """
+        Validate model credentials
+
+        :param model: model name
+        :param credentials: model credentials
+        :return:
+        """
+        try:
+            self.invoke(
+                model=model,
+                credentials=credentials,
+                query="What is the capital of the United States?",
+                docs=[
+                    "Carson City is the capital city of the American state of Nevada. At the 2010 United States "
+                    "Census, Carson City had a population of 55,274.",
+                    "The Commonwealth of the Northern Mariana Islands is a group of islands in the Pacific Ocean that "
+                    "are a political division controlled by the United States. Its capital is Saipan.",
+                ],
+                score_threshold=0.8,
+            )
+        except Exception as ex:
+            print(ex)
+            raise CredentialsValidateFailedError(str(ex))
+
+    @property
+    def _invoke_error_mapping(self) -> dict[type[InvokeError], list[type[Exception]]]:
+        """
+        Map model invoke error to unified error
+        The key is the error type thrown to the caller
+        The value is the error type thrown by the model,
+        which needs to be converted into a unified error type for the caller.
+
+        :return: Invoke error mapping
+        """
+        return {
+            InvokeConnectionError: [
+                RequestFailure,
+            ],
+            InvokeServerUnavailableError: [
+                ServiceUnavailableError,
+            ],
+            InvokeRateLimitError: [],
+            InvokeAuthorizationError: [
+                AuthenticationError,
+            ],
+            InvokeBadRequestError: [
+                InvalidParameter,
+                UnsupportedModel,
+                UnsupportedHTTPMethod,
+            ],
+        }

api/core/model_runtime/model_providers/tongyi/tongyi.yaml

@@ -18,6 +18,7 @@ supported_model_types:
   - llm
   - tts
   - text-embedding
+  - rerank
 configurate_methods:
   - predefined-model
   - customizable-model

api/core/model_runtime/model_providers/tongyi/tts/tts.py

@@ -1,6 +1,6 @@
 import threading
 from queue import Queue
-from typing import Optional
+from typing import Any, Optional
 
 import dashscope
 from dashscope import SpeechSynthesizer
@@ -20,7 +20,7 @@ class TongyiText2SpeechModel(_CommonTongyi, TTSModel):
 
     def _invoke(
         self, model: str, tenant_id: str, credentials: dict, content_text: str, voice: str, user: Optional[str] = None
-    ) -> any:
+    ) -> Any:
         """
         _invoke text2speech model
 
@@ -58,7 +58,7 @@ class TongyiText2SpeechModel(_CommonTongyi, TTSModel):
         except Exception as ex:
             raise CredentialsValidateFailedError(str(ex))
 
-    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> any:
+    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> Any:
         """
         _tts_invoke_streaming text2speech model
 

api/core/model_runtime/model_providers/vertex_ai/llm/llm.py

@@ -7,6 +7,7 @@ from collections.abc import Generator
 from typing import Optional, Union, cast
 
 import google.auth.transport.requests
+import requests
 import vertexai.generative_models as glm
 from anthropic import AnthropicVertex, Stream
 from anthropic.types import (
@@ -653,9 +654,15 @@ class VertexAiLargeLanguageModel(LargeLanguageModel):
                     if c.type == PromptMessageContentType.TEXT:
                         parts.append(glm.Part.from_text(c.data))
                     else:
-                        metadata, data = c.data.split(",", 1)
+                        message_content = cast(ImagePromptMessageContent, c)
-                        mime_type = metadata.split(";", 1)[0].split(":")[1]
+                        if not message_content.data.startswith("data:"):
-                        parts.append(glm.Part.from_data(mime_type=mime_type, data=data))
+                            url_arr = message_content.data.split(".")
+                            mime_type = f"image/{url_arr[-1]}"
+                            parts.append(glm.Part.from_uri(mime_type=mime_type, uri=message_content.data))
+                        else:
+                            metadata, data = c.data.split(",", 1)
+                            mime_type = metadata.split(";", 1)[0].split(":")[1]
+                            parts.append(glm.Part.from_data(mime_type=mime_type, data=data))
                 glm_content = glm.Content(role="user", parts=parts)
             return glm_content
         elif isinstance(message, AssistantPromptMessage):

api/core/model_runtime/model_providers/wenxin/llm/llm.py

@@ -64,7 +64,7 @@ class ErnieBotLargeLanguageModel(LargeLanguageModel):
         stop: Optional[list[str]] = None,
         stream: bool = True,
         user: Optional[str] = None,
-        callbacks: list[Callback] = None,
+        callbacks: Optional[list[Callback]] = None,
     ) -> Union[LLMResult, Generator]:
         """
         Code block mode wrapper for invoking large language model

api/core/model_runtime/model_providers/xinference/tts/tts.py

@@ -1,5 +1,5 @@
 import concurrent.futures
-from typing import Optional
+from typing import Any, Optional
 
 from xinference_client.client.restful.restful_client import RESTfulAudioModelHandle
 
@@ -166,7 +166,7 @@ class XinferenceText2SpeechModel(TTSModel):
 
         return self.model_voices["__default"]["all"]
 
-    def _get_model_default_voice(self, model: str, credentials: dict) -> any:
+    def _get_model_default_voice(self, model: str, credentials: dict) -> Any:
         return ""
 
     def _get_model_word_limit(self, model: str, credentials: dict) -> int:
@@ -178,7 +178,7 @@ class XinferenceText2SpeechModel(TTSModel):
     def _get_model_workers_limit(self, model: str, credentials: dict) -> int:
         return 5
 
-    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> any:
+    def _tts_invoke_streaming(self, model: str, credentials: dict, content_text: str, voice: str) -> Any:
         """
         _tts_invoke_streaming text2speech model
 

api/core/model_runtime/model_providers/zhipuai/llm/llm.py

@@ -223,6 +223,16 @@ class ZhipuAILargeLanguageModel(_CommonZhipuaiAI, LargeLanguageModel):
                 else:
                     new_prompt_messages.append(copy_prompt_message)
 
+        # zhipuai moved web_search param to tools
+        if "web_search" in model_parameters:
+            enable_web_search = model_parameters.get("web_search")
+            model_parameters.pop("web_search")
+            web_search_params = {"type": "web_search", "web_search": {"enable": enable_web_search}}
+            if "tools" in model_parameters:
+                model_parameters["tools"].append(web_search_params)
+            else:
+                model_parameters["tools"] = [web_search_params]
+
         if model in {"glm-4v", "glm-4v-plus"}:
             params = self._construct_glm_4v_parameter(model, new_prompt_messages, model_parameters)
         else:

api/core/model_runtime/model_providers/zhipuai/zhipuai_sdk/api_resource/assistant/assistant.py

@@ -41,8 +41,8 @@ class Assistant(BaseAPI):
         conversation_id: Optional[str] = None,
         attachments: Optional[list[assistant_create_params.AssistantAttachments]] = None,
         metadata: dict | None = None,
-        request_id: str = None,
+        request_id: Optional[str] = None,
-        user_id: str = None,
+        user_id: Optional[str] = None,
         extra_headers: Headers | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
@@ -72,9 +72,9 @@ class Assistant(BaseAPI):
     def query_support(
         self,
         *,
-        assistant_id_list: list[str] = None,
+        assistant_id_list: Optional[list[str]] = None,
-        request_id: str = None,
+        request_id: Optional[str] = None,
-        user_id: str = None,
+        user_id: Optional[str] = None,
         extra_headers: Headers | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
@@ -99,8 +99,8 @@ class Assistant(BaseAPI):
         page: int = 1,
         page_size: int = 10,
         *,
-        request_id: str = None,
+        request_id: Optional[str] = None,
-        user_id: str = None,
+        user_id: Optional[str] = None,
         extra_headers: Headers | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,

api/core/model_runtime/model_providers/zhipuai/zhipuai_sdk/api_resource/files.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 
 from collections.abc import Mapping
-from typing import TYPE_CHECKING, Literal, cast
+from typing import TYPE_CHECKING, Literal, Optional, cast
 
 import httpx
 
@@ -34,11 +34,11 @@ class Files(BaseAPI):
     def create(
         self,
         *,
-        file: FileTypes = None,
+        file: Optional[FileTypes] = None,
-        upload_detail: list[UploadDetail] = None,
+        upload_detail: Optional[list[UploadDetail]] = None,
         purpose: Literal["fine-tune", "retrieval", "batch"],
-        knowledge_id: str = None,
+        knowledge_id: Optional[str] = None,
-        sentence_size: int = None,
+        sentence_size: Optional[int] = None,
         extra_headers: Headers | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,

api/core/model_runtime/model_providers/zhipuai/zhipuai_sdk/api_resource/knowledge/document/document.py

@@ -34,12 +34,12 @@ class Document(BaseAPI):
     def create(
         self,
         *,
-        file: FileTypes = None,
+        file: Optional[FileTypes] = None,
         custom_separator: Optional[list[str]] = None,
-        upload_detail: list[UploadDetail] = None,
+        upload_detail: Optional[list[UploadDetail]] = None,
         purpose: Literal["retrieval"],
-        knowledge_id: str = None,
+        knowledge_id: Optional[str] = None,
-        sentence_size: int = None,
+        sentence_size: Optional[int] = None,
         extra_headers: Headers | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,

api/core/model_runtime/model_providers/zhipuai/zhipuai_sdk/api_resource/videos/videos.py

@@ -31,11 +31,11 @@ class Videos(BaseAPI):
         self,
         model: str,
         *,
-        prompt: str = None,
+        prompt: Optional[str] = None,
-        image_url: str = None,
+        image_url: Optional[str] = None,
         sensitive_word_check: Optional[SensitiveWordCheckRequest] | NotGiven = NOT_GIVEN,
-        request_id: str = None,
+        request_id: Optional[str] = None,
-        user_id: str = None,
+        user_id: Optional[str] = None,
         extra_headers: Headers | None = None,
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,

api/core/rag/datasource/vdb/elasticsearch/elasticsearch_vector.py

@@ -1,5 +1,6 @@
 import json
 import logging
+import math
 from typing import Any, Optional
 from urllib.parse import urlparse
 
@@ -112,7 +113,8 @@ class ElasticSearchVector(BaseVector):
 
     def search_by_vector(self, query_vector: list[float], **kwargs: Any) -> list[Document]:
         top_k = kwargs.get("top_k", 10)
-        knn = {"field": Field.VECTOR.value, "query_vector": query_vector, "k": top_k}
+        num_candidates = math.ceil(top_k * 1.5)
+        knn = {"field": Field.VECTOR.value, "query_vector": query_vector, "k": top_k, "num_candidates": num_candidates}
 
         results = self._client.search(index=self._collection_name, knn=knn, size=top_k)
 

api/core/rag/datasource/vdb/relyt/relyt_vector.py

@@ -162,7 +162,7 @@ class RelytVector(BaseVector):
         else:
             return None
 
-    def delete_by_uuids(self, ids: list[str] = None):
+    def delete_by_uuids(self, ids: Optional[list[str]] = None):
         """Delete by vector IDs.
 
         Args:

api/core/rag/datasource/vdb/vector_factory.py

@@ -1,5 +1,5 @@
 from abc import ABC, abstractmethod
-from typing import Any
+from typing import Any, Optional
 
 from configs import dify_config
 from core.embedding.cached_embedding import CacheEmbedding
@@ -25,7 +25,7 @@ class AbstractVectorFactory(ABC):
 
 
 class Vector:
-    def __init__(self, dataset: Dataset, attributes: list = None):
+    def __init__(self, dataset: Dataset, attributes: Optional[list] = None):
         if attributes is None:
             attributes = ["doc_id", "dataset_id", "document_id", "doc_hash"]
         self._dataset = dataset
@@ -106,7 +106,7 @@ class Vector:
             case _:
                 raise ValueError(f"Vector store {vector_type} is not supported.")
 
-    def create(self, texts: list = None, **kwargs):
+    def create(self, texts: Optional[list] = None, **kwargs):
         if texts:
             embeddings = self._embeddings.embed_documents([document.page_content for document in texts])
             self._vector_processor.create(texts=texts, embeddings=embeddings, **kwargs)

api/core/rag/entities/context_entities.py

@@ -1,3 +1,5 @@
+from typing import Optional
+
 from pydantic import BaseModel
 
 
@@ -7,4 +9,4 @@ class DocumentContext(BaseModel):
     """
 
     content: str
-    score: float
+    score: Optional[float] = None

api/core/rag/extractor/extract_processor.py

@@ -1,7 +1,7 @@
 import re
 import tempfile
 from pathlib import Path
-from typing import Union
+from typing import Optional, Union
 from urllib.parse import unquote
 
 from configs import dify_config
@@ -84,7 +84,7 @@ class ExtractProcessor:
 
     @classmethod
     def extract(
-        cls, extract_setting: ExtractSetting, is_automatic: bool = False, file_path: str = None
+        cls, extract_setting: ExtractSetting, is_automatic: bool = False, file_path: Optional[str] = None
     ) -> list[Document]:
         if extract_setting.datasource_type == DatasourceType.FILE.value:
             with tempfile.TemporaryDirectory() as temp_dir:

api/core/rag/extractor/unstructured/unstructured_epub_extractor.py

@@ -1,4 +1,5 @@
 import logging
+from typing import Optional
 
 from core.rag.extractor.extractor_base import BaseExtractor
 from core.rag.models.document import Document
@@ -17,7 +18,7 @@ class UnstructuredEpubExtractor(BaseExtractor):
     def __init__(
         self,
         file_path: str,
-        api_url: str = None,
+        api_url: Optional[str] = None,
     ):
         """Initialize with file path."""
         self._file_path = file_path

api/core/rag/retrieval/dataset_retrieval.py

@@ -231,6 +231,9 @@ class DatasetRetrieval:
                                 source["content"] = segment.content
                             retrieval_resource_list.append(source)
         if hit_callback and retrieval_resource_list:
+            retrieval_resource_list = sorted(retrieval_resource_list, key=lambda x: x.get("score"), reverse=True)
+            for position, item in enumerate(retrieval_resource_list, start=1):
+                item["position"] = position
             hit_callback.return_retriever_resource_info(retrieval_resource_list)
         if document_context_list:
             document_context_list = sorted(document_context_list, key=lambda x: x.score, reverse=True)
@@ -536,7 +539,7 @@ class DatasetRetrieval:
                 continue
 
             # pass if dataset is not available
-            if dataset and dataset.available_document_count == 0:
+            if dataset and dataset.provider != "external" and dataset.available_document_count == 0:
                 continue
 
             available_datasets.append(dataset)

api/core/tools/entities/tool_entities.py

@@ -341,7 +341,7 @@ class ToolRuntimeVariablePool(BaseModel):
 
         self.pool.append(variable)
 
-    def set_file(self, tool_name: str, value: str, name: str = None) -> None:
+    def set_file(self, tool_name: str, value: str, name: Optional[str] = None) -> None:
         """
         set an image variable
 

api/core/tools/provider/_position.yaml

@@ -5,31 +5,41 @@
 - searchapi
 - serper
 - searxng
+- websearch
+- tavily
+- stackexchange
+- pubmed
+- arxiv
+- aws
+- nominatim
+- devdocs
+- spider
+- firecrawl
+- brave
+- crossref
+- jina
+- webscraper
 - dalle
 - azuredalle
 - stability
-- wikipedia
-- nominatim
-- yahoo
-- alphavantage
-- arxiv
-- pubmed
 - stablediffusion
-- webscraper
+- cogview
-- jina
+- comfyui
+- getimgai
+- siliconflow
+- spark
+- stepfun
+- xinference
+- alphavantage
+- yahoo
+- openweather
+- gaode
 - aippt
-- youtube
-- code
-- wolframalpha
-- maths
-- github
 - chart
-- time
+- youtube
-- vectorizer
+- did
-- gaode
-- wecom
-- qrcode
 - dingtalk
+- discord
 - feishu
 - feishu_base
 - feishu_document
@@ -39,4 +49,24 @@
 - feishu_calendar
 - feishu_spreadsheet
 - slack
+- twilio
+- wecom
+- wikipedia
+- code
+- wolframalpha
+- maths
+- github
+- gitlab
+- time
+- vectorizer
+- qrcode
 - tianditu
+- google_translate
+- hap
+- json_process
+- judge0ce
+- novitaai
+- onebot
+- regex
+- trello
+- vanna

api/core/tools/provider/builtin/aws/tools/sagemaker_tts.py

@@ -1,6 +1,6 @@
 import json
 from enum import Enum
-from typing import Any, Union
+from typing import Any, Optional, Union
 
 import boto3
 
@@ -21,7 +21,7 @@ class SageMakerTTSTool(BuiltinTool):
     s3_client: Any = None
     comprehend_client: Any = None
 
-    def _detect_lang_code(self, content: str, map_dict: dict = None):
+    def _detect_lang_code(self, content: str, map_dict: Optional[dict] = None):
         map_dict = {"zh": "<|zh|>", "en": "<|en|>", "ja": "<|jp|>", "zh-TW": "<|yue|>", "ko": "<|ko|>"}
 
         response = self.comprehend_client.detect_dominant_language(Text=content)

api/core/tools/provider/builtin/jina/jina.yaml

@@ -6,9 +6,9 @@ identity:
     zh_Hans: Jina AI
     pt_BR: Jina AI
   description:
-    en_US: Convert any URL to an LLM-friendly input or perform searches on the web for grounding information. Experience improved output for your agent and RAG systems at no cost.
+    en_US: Your Search Foundation, Supercharged!
-    zh_Hans: 将任何URL转换为LLM易读的输入或在网页上搜索引擎上搜索引擎。
+    zh_Hans: 您的搜索底座，从此不同!
-    pt_BR: Converte qualquer URL em uma entrada LLm-fácil de ler ou realize pesquisas na web para obter informação de grounding. Tenha uma experiência melhor para seu agente e sistemas RAG sem custo.
+    pt_BR: Your Search Foundation, Supercharged!
   icon: icon.svg
   tags:
     - search

api/core/tools/provider/builtin/time/tools/localtime_to_timestamp.py

@@ -0,0 +1,44 @@
+from datetime import datetime
+from typing import Any, Union
+
+import pytz
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.errors import ToolInvokeError
+from core.tools.tool.builtin_tool import BuiltinTool
+
+
+class LocaltimeToTimestampTool(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+    ) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
+        """
+        Convert localtime to timestamp
+        """
+        localtime = tool_parameters.get("localtime")
+        timezone = tool_parameters.get("timezone", "Asia/Shanghai")
+        if not timezone:
+            timezone = None
+        time_format = "%Y-%m-%d %H:%M:%S"
+
+        timestamp = self.localtime_to_timestamp(localtime, time_format, timezone)
+        if not timestamp:
+            return self.create_text_message(f"Invalid localtime: {localtime}")
+
+        return self.create_text_message(f"{timestamp}")
+
+    @staticmethod
+    def localtime_to_timestamp(localtime: str, time_format: str, local_tz=None) -> int | None:
+        try:
+            if local_tz is None:
+                local_tz = datetime.now().astimezone().tzinfo
+            if isinstance(local_tz, str):
+                local_tz = pytz.timezone(local_tz)
+            local_time = datetime.strptime(localtime, time_format)
+            localtime = local_tz.localize(local_time)
+            timestamp = int(localtime.timestamp())
+            return timestamp
+        except Exception as e:
+            raise ToolInvokeError(str(e))

api/core/tools/provider/builtin/time/tools/localtime_to_timestamp.yaml

@@ -0,0 +1,33 @@
+identity:
+  name: localtime_to_timestamp
+  author: zhuhao
+  label:
+    en_US: localtime to timestamp
+    zh_Hans: 获取时间戳
+description:
+  human:
+    en_US: A tool for localtime convert to timestamp
+    zh_Hans: 获取时间戳
+  llm: A tool for localtime convert to timestamp
+parameters:
+  - name: localtime
+    type: string
+    required: true
+    form: llm
+    label:
+      en_US: localtime
+      zh_Hans: 本地时间
+    human_description:
+      en_US: localtime, such as 2024-1-1 0:0:0
+      zh_Hans: 本地时间, 比如2024-1-1 0:0:0
+  - name: timezone
+    type: string
+    required: false
+    form: llm
+    label:
+      en_US: Timezone
+      zh_Hans: 时区
+    human_description:
+      en_US: Timezone, such as Asia/Shanghai
+      zh_Hans: 时区, 比如Asia/Shanghai
+    default: Asia/Shanghai

api/core/tools/provider/builtin/time/tools/timestamp_to_localtime.py

@@ -0,0 +1,44 @@
+from datetime import datetime
+from typing import Any, Union
+
+import pytz
+
+from core.tools.entities.tool_entities import ToolInvokeMessage
+from core.tools.errors import ToolInvokeError
+from core.tools.tool.builtin_tool import BuiltinTool
+
+
+class TimestampToLocaltimeTool(BuiltinTool):
+    def _invoke(
+        self,
+        user_id: str,
+        tool_parameters: dict[str, Any],
+    ) -> Union[ToolInvokeMessage, list[ToolInvokeMessage]]:
+        """
+        Convert timestamp to localtime
+        """
+        timestamp = tool_parameters.get("timestamp")
+        timezone = tool_parameters.get("timezone", "Asia/Shanghai")
+        if not timezone:
+            timezone = None
+        time_format = "%Y-%m-%d %H:%M:%S"
+
+        locatime = self.timestamp_to_localtime(timestamp, timezone)
+        if not locatime:
+            return self.create_text_message(f"Invalid timestamp: {timestamp}")
+
+        localtime_format = locatime.strftime(time_format)
+
+        return self.create_text_message(f"{localtime_format}")
+
+    @staticmethod
+    def timestamp_to_localtime(timestamp: int, local_tz=None) -> datetime | None:
+        try:
+            if local_tz is None:
+                local_tz = datetime.now().astimezone().tzinfo
+            if isinstance(local_tz, str):
+                local_tz = pytz.timezone(local_tz)
+            local_time = datetime.fromtimestamp(timestamp, local_tz)
+            return local_time
+        except Exception as e:
+            raise ToolInvokeError(str(e))

api/core/tools/provider/builtin/time/tools/timestamp_to_localtime.yaml

@@ -0,0 +1,33 @@
+identity:
+  name: timestamp_to_localtime
+  author: zhuhao
+  label:
+    en_US: Timestamp to localtime
+    zh_Hans: 时间戳转换
+description:
+  human:
+    en_US: A tool for timestamp convert to localtime
+    zh_Hans: 时间戳转换
+  llm: A tool for timestamp convert to localtime
+parameters:
+  - name: timestamp
+    type: number
+    required: true
+    form: llm
+    label:
+      en_US: Timestamp
+      zh_Hans: 时间戳
+    human_description:
+      en_US: Timestamp
+      zh_Hans: 时间戳
+  - name: timezone
+    type: string
+    required: false
+    form: llm
+    label:
+      en_US: Timezone
+      zh_Hans: 时区
+    human_description:
+      en_US: Timezone, such as Asia/Shanghai
+      zh_Hans: 时区, 比如Asia/Shanghai
+    default: Asia/Shanghai

api/core/tools/provider/builtin/vanna/tools/vanna.py

@@ -111,9 +111,10 @@ class VannaTool(BuiltinTool):
         # with "visualize" set to True (default behavior) leads to remote code execution.
         # Affected versions: <= 0.5.5
         #########################################################################################
-        generate_chart = False
+        allow_llm_to_see_data = tool_parameters.get("allow_llm_to_see_data", False)
-        # generate_chart = tool_parameters.get("generate_chart", True)
+        res = vn.ask(
-        res = vn.ask(prompt, False, True, generate_chart)
+            prompt, print_results=False, auto_train=True, visualize=False, allow_llm_to_see_data=allow_llm_to_see_data
+        )
 
         result = []
 

api/core/tools/provider/builtin/vanna/tools/vanna.yaml

@@ -200,14 +200,14 @@ parameters:
       en_US: If enabled, it will attempt to train on the metadata of that database
       zh_Hans: 是否自动从数据库获取元数据来训练
     form: form
-  - name: generate_chart
+  - name: allow_llm_to_see_data
     type: boolean
     required: false
-    default: True
+    default: false
     label:
-      en_US: Generate Charts
+      en_US: Whether to allow the LLM to see the data
-      zh_Hans: 生成图表
+      zh_Hans: 是否允许LLM查看数据
     human_description:
-      en_US: Generate Charts
+      en_US: Whether to allow the LLM to see the data
-      zh_Hans: 是否生成图表
+      zh_Hans: 是否允许LLM查看数据
     form: form

api/core/tools/provider/builtin/vanna/vanna.yaml

@@ -8,6 +8,9 @@ identity:
     en_US: The fastest way to get actionable insights from your database just by asking questions.
     zh_Hans: 一个基于大模型和RAG的Text2SQL工具。
   icon: icon.png
+  tags:
+    - utilities
+    - productivity
 credentials_for_provider:
   api_key:
     type: secret-input

api/core/tools/provider/builtin/xinference/tools/stable_diffusion.py

@@ -104,14 +104,15 @@ class StableDiffusionTool(BuiltinTool):
         model = self.runtime.credentials.get("model", None)
         if not model:
             return self.create_text_message("Please input model")
-
+        api_key = self.runtime.credentials.get("api_key") or "abc"
+        headers = {"Authorization": f"Bearer {api_key}"}
         # set model
         try:
             url = str(URL(base_url) / "sdapi" / "v1" / "options")
             response = post(
                 url,
                 json={"sd_model_checkpoint": model},
-                headers={"Authorization": f"Bearer {self.runtime.credentials['api_key']}"},
+                headers=headers,
             )
             if response.status_code != 200:
                 raise ToolProviderCredentialValidationError("Failed to set model, please tell user to set model")
@@ -257,14 +258,15 @@ class StableDiffusionTool(BuiltinTool):
             draw_options["prompt"] = f"{lora},{prompt}"
         else:
             draw_options["prompt"] = prompt
-
+        api_key = self.runtime.credentials.get("api_key") or "abc"
+        headers = {"Authorization": f"Bearer {api_key}"}
         try:
             url = str(URL(base_url) / "sdapi" / "v1" / "img2img")
             response = post(
                 url,
                 json=draw_options,
                 timeout=120,
-                headers={"Authorization": f"Bearer {self.runtime.credentials['api_key']}"},
+                headers=headers,
             )
             if response.status_code != 200:
                 return self.create_text_message("Failed to generate image")
@@ -298,14 +300,15 @@ class StableDiffusionTool(BuiltinTool):
         else:
             draw_options["prompt"] = prompt
         draw_options["override_settings"]["sd_model_checkpoint"] = model
-
+        api_key = self.runtime.credentials.get("api_key") or "abc"
+        headers = {"Authorization": f"Bearer {api_key}"}
         try:
             url = str(URL(base_url) / "sdapi" / "v1" / "txt2img")
             response = post(
                 url,
                 json=draw_options,
                 timeout=120,
-                headers={"Authorization": f"Bearer {self.runtime.credentials['api_key']}"},
+                headers=headers,
             )
             if response.status_code != 200:
                 return self.create_text_message("Failed to generate image")

api/core/tools/provider/builtin/xinference/xinference.py

@@ -6,12 +6,18 @@ from core.tools.provider.builtin_tool_provider import BuiltinToolProviderControl
 
 class XinferenceProvider(BuiltinToolProviderController):
     def _validate_credentials(self, credentials: dict) -> None:
-        base_url = credentials.get("base_url")
+        base_url = credentials.get("base_url", "").removesuffix("/")
-        api_key = credentials.get("api_key")
+        api_key = credentials.get("api_key", "")
-        model = credentials.get("model")
+        if not api_key:
+            api_key = "abc"
+            credentials["api_key"] = api_key
+        model = credentials.get("model", "")
+        if not base_url or not model:
+            raise ToolProviderCredentialValidationError("Xinference base_url and model is required")
+        headers = {"Authorization": f"Bearer {api_key}"}
         res = requests.post(
             f"{base_url}/sdapi/v1/options",
-            headers={"Authorization": f"Bearer {api_key}"},
+            headers=headers,
             json={"sd_model_checkpoint": model},
         )
         if res.status_code != 200:

api/core/tools/provider/builtin/xinference/xinference.yaml

@@ -31,7 +31,7 @@ credentials_for_provider:
       zh_Hans: 请输入你的模型名称
   api_key:
     type: secret-input
-    required: true
+    required: false
     label:
       en_US: API Key
       zh_Hans: Xinference 服务器的 API Key

api/core/tools/tool/builtin_tool.py

@@ -1,3 +1,5 @@
+from typing import Optional
+
 from core.model_runtime.entities.llm_entities import LLMResult
 from core.model_runtime.entities.message_entities import PromptMessage, SystemPromptMessage, UserPromptMessage
 from core.tools.entities.tool_entities import ToolProviderType
@@ -124,7 +126,7 @@ class BuiltinTool(Tool):
 
         return result
 
-    def get_url(self, url: str, user_agent: str = None) -> str:
+    def get_url(self, url: str, user_agent: Optional[str] = None) -> str:
         """
         get url
         """

api/core/tools/tool/dataset_retriever/dataset_retriever_tool.py

@@ -1,10 +1,12 @@
 from pydantic import BaseModel, Field
 
 from core.rag.datasource.retrieval_service import RetrievalService
+from core.rag.models.document import Document as RetrievalDocument
 from core.rag.retrieval.retrieval_methods import RetrievalMethod
 from core.tools.tool.dataset_retriever.dataset_retriever_base_tool import DatasetRetrieverBaseTool
 from extensions.ext_database import db
 from models.dataset import Dataset, Document, DocumentSegment
+from services.external_knowledge_service import ExternalDatasetService
 
 default_retrieval_model = {
     "search_method": RetrievalMethod.SEMANTIC_SEARCH.value,
@@ -53,97 +55,137 @@ class DatasetRetrieverTool(DatasetRetrieverBaseTool):
 
         for hit_callback in self.hit_callbacks:
             hit_callback.on_query(query, dataset.id)
-
+        if dataset.provider == "external":
-        # get retrieval model , if the model is not setting , using default
+            results = []
-        retrieval_model = dataset.retrieval_model or default_retrieval_model
+            external_documents = ExternalDatasetService.fetch_external_knowledge_retrieval(
-        if dataset.indexing_technique == "economy":
+                tenant_id=dataset.tenant_id,
-            # use keyword table query
+                dataset_id=dataset.id,
-            documents = RetrievalService.retrieve(
+                query=query,
-                retrieval_method="keyword_search", dataset_id=dataset.id, query=query, top_k=self.top_k
+                external_retrieval_parameters=dataset.retrieval_model,
             )
-            return str("\n".join([document.page_content for document in documents]))
+            for external_document in external_documents:
+                document = RetrievalDocument(
+                    page_content=external_document.get("content"),
+                    metadata=external_document.get("metadata"),
+                    provider="external",
+                )
+                document.metadata["score"] = external_document.get("score")
+                document.metadata["title"] = external_document.get("title")
+                document.metadata["dataset_id"] = dataset.id
+                document.metadata["dataset_name"] = dataset.name
+                results.append(document)
+            # deal with external documents
+            context_list = []
+            for position, item in enumerate(results, start=1):
+                source = {
+                    "position": position,
+                    "dataset_id": item.metadata.get("dataset_id"),
+                    "dataset_name": item.metadata.get("dataset_name"),
+                    "document_name": item.metadata.get("title"),
+                    "data_source_type": "external",
+                    "retriever_from": self.retriever_from,
+                    "score": item.metadata.get("score"),
+                    "title": item.metadata.get("title"),
+                    "content": item.page_content,
+                }
+                context_list.append(source)
+            for hit_callback in self.hit_callbacks:
+                hit_callback.return_retriever_resource_info(context_list)
+
+            return str("\n".join([item.page_content for item in results]))
         else:
-            if self.top_k > 0:
+            # get retrieval model , if the model is not setting , using default
-                # retrieval source
+            retrieval_model = dataset.retrieval_model or default_retrieval_model
+            if dataset.indexing_technique == "economy":
+                # use keyword table query
                 documents = RetrievalService.retrieve(
-                    retrieval_method=retrieval_model.get("search_method", "semantic_search"),
+                    retrieval_method="keyword_search", dataset_id=dataset.id, query=query, top_k=self.top_k
-                    dataset_id=dataset.id,
-                    query=query,
-                    top_k=self.top_k,
-                    score_threshold=retrieval_model.get("score_threshold", 0.0)
-                    if retrieval_model["score_threshold_enabled"]
-                    else 0.0,
-                    reranking_model=retrieval_model.get("reranking_model", None)
-                    if retrieval_model["reranking_enable"]
-                    else None,
-                    reranking_mode=retrieval_model.get("reranking_mode") or "reranking_model",
-                    weights=retrieval_model.get("weights", None),
                 )
+                return str("\n".join([document.page_content for document in documents]))
             else:
-                documents = []
+                if self.top_k > 0:
-
+                    # retrieval source
-            for hit_callback in self.hit_callbacks:
+                    documents = RetrievalService.retrieve(
-                hit_callback.on_tool_end(documents)
+                        retrieval_method=retrieval_model.get("search_method", "semantic_search"),
-            document_score_list = {}
+                        dataset_id=dataset.id,
-            if dataset.indexing_technique != "economy":
+                        query=query,
-                for item in documents:
+                        top_k=self.top_k,
-                    if item.metadata.get("score"):
+                        score_threshold=retrieval_model.get("score_threshold", 0.0)
-                        document_score_list[item.metadata["doc_id"]] = item.metadata["score"]
+                        if retrieval_model["score_threshold_enabled"]
-            document_context_list = []
+                        else 0.0,
-            index_node_ids = [document.metadata["doc_id"] for document in documents]
+                        reranking_model=retrieval_model.get("reranking_model", None)
-            segments = DocumentSegment.query.filter(
+                        if retrieval_model["reranking_enable"]
-                DocumentSegment.dataset_id == self.dataset_id,
+                        else None,
-                DocumentSegment.completed_at.isnot(None),
+                        reranking_mode=retrieval_model.get("reranking_mode") or "reranking_model",
-                DocumentSegment.status == "completed",
+                        weights=retrieval_model.get("weights", None),
-                DocumentSegment.enabled == True,
+                    )
-                DocumentSegment.index_node_id.in_(index_node_ids),
+                else:
-            ).all()
+                    documents = []
-
+
-            if segments:
+                for hit_callback in self.hit_callbacks:
-                index_node_id_to_position = {id: position for position, id in enumerate(index_node_ids)}
+                    hit_callback.on_tool_end(documents)
-                sorted_segments = sorted(
+                document_score_list = {}
-                    segments, key=lambda segment: index_node_id_to_position.get(segment.index_node_id, float("inf"))
+                if dataset.indexing_technique != "economy":
-                )
+                    for item in documents:
-                for segment in sorted_segments:
+                        if item.metadata.get("score"):
-                    if segment.answer:
+                            document_score_list[item.metadata["doc_id"]] = item.metadata["score"]
-                        document_context_list.append(f"question:{segment.get_sign_content()} answer:{segment.answer}")
+                document_context_list = []
-                    else:
+                index_node_ids = [document.metadata["doc_id"] for document in documents]
-                        document_context_list.append(segment.get_sign_content())
+                segments = DocumentSegment.query.filter(
-                if self.return_resource:
+                    DocumentSegment.dataset_id == self.dataset_id,
-                    context_list = []
+                    DocumentSegment.completed_at.isnot(None),
-                    resource_number = 1
+                    DocumentSegment.status == "completed",
+                    DocumentSegment.enabled == True,
+                    DocumentSegment.index_node_id.in_(index_node_ids),
+                ).all()
+
+                if segments:
+                    index_node_id_to_position = {id: position for position, id in enumerate(index_node_ids)}
+                    sorted_segments = sorted(
+                        segments, key=lambda segment: index_node_id_to_position.get(segment.index_node_id, float("inf"))
+                    )
                     for segment in sorted_segments:
-                        context = {}
+                        if segment.answer:
-                        document = Document.query.filter(
+                            document_context_list.append(
-                            Document.id == segment.document_id,
+                                f"question:{segment.get_sign_content()} answer:{segment.answer}"
-                            Document.enabled == True,
+                            )
-                            Document.archived == False,
+                        else:
-                        ).first()
+                            document_context_list.append(segment.get_sign_content())
-                        if dataset and document:
+                    if self.return_resource:
-                            source = {
+                        context_list = []
-                                "position": resource_number,
+                        resource_number = 1
-                                "dataset_id": dataset.id,
+                        for segment in sorted_segments:
-                                "dataset_name": dataset.name,
+                            context = {}
-                                "document_id": document.id,
+                            document = Document.query.filter(
-                                "document_name": document.name,
+                                Document.id == segment.document_id,
-                                "data_source_type": document.data_source_type,
+                                Document.enabled == True,
-                                "segment_id": segment.id,
+                                Document.archived == False,
-                                "retriever_from": self.retriever_from,
+                            ).first()
-                                "score": document_score_list.get(segment.index_node_id, None),
+                            if dataset and document:
-                            }
+                                source = {
-                            if self.retriever_from == "dev":
+                                    "position": resource_number,
-                                source["hit_count"] = segment.hit_count
+                                    "dataset_id": dataset.id,
-                                source["word_count"] = segment.word_count
+                                    "dataset_name": dataset.name,
-                                source["segment_position"] = segment.position
+                                    "document_id": document.id,
-                                source["index_node_hash"] = segment.index_node_hash
+                                    "document_name": document.name,
-                            if segment.answer:
+                                    "data_source_type": document.data_source_type,
-                                source["content"] = f"question:{segment.content} \nanswer:{segment.answer}"
+                                    "segment_id": segment.id,
-                            else:
+                                    "retriever_from": self.retriever_from,
-                                source["content"] = segment.content
+                                    "score": document_score_list.get(segment.index_node_id, None),
-                            context_list.append(source)
+                                }
-                        resource_number += 1
+                                if self.retriever_from == "dev":
-
+                                    source["hit_count"] = segment.hit_count
-                    for hit_callback in self.hit_callbacks:
+                                    source["word_count"] = segment.word_count
-                        hit_callback.return_retriever_resource_info(context_list)
+                                    source["segment_position"] = segment.position
-
+                                    source["index_node_hash"] = segment.index_node_hash
-            return str("\n".join(document_context_list))
+                                if segment.answer:
+                                    source["content"] = f"question:{segment.content} \nanswer:{segment.answer}"
+                                else:
+                                    source["content"] = segment.content
+                                context_list.append(source)
+                            resource_number += 1
+
+                        for hit_callback in self.hit_callbacks:
+                            hit_callback.return_retriever_resource_info(context_list)
+
+                return str("\n".join(document_context_list))

api/core/tools/tool/tool.py

@@ -318,7 +318,7 @@ class Tool(BaseModel, ABC):
         """
         return ToolInvokeMessage(type=ToolInvokeMessage.MessageType.TEXT, message=text, save_as=save_as)
 
-    def create_blob_message(self, blob: bytes, meta: dict = None, save_as: str = "") -> ToolInvokeMessage:
+    def create_blob_message(self, blob: bytes, meta: Optional[dict] = None, save_as: str = "") -> ToolInvokeMessage:
         """
         create a blob message
 

api/core/tools/tool_manager.py

@@ -4,7 +4,7 @@ import mimetypes
 from collections.abc import Generator
 from os import listdir, path
 from threading import Lock
-from typing import Any, Union
+from typing import Any, Optional, Union
 
 from configs import dify_config
 from core.agent.entities import AgentToolEntity
@@ -72,7 +72,7 @@ class ToolManager:
 
     @classmethod
     def get_tool(
-        cls, provider_type: str, provider_id: str, tool_name: str, tenant_id: str = None
+        cls, provider_type: str, provider_id: str, tool_name: str, tenant_id: Optional[str] = None
     ) -> Union[BuiltinTool, ApiTool]:
         """
         get the tool

api/core/tools/utils/feishu_api_utils.py

@@ -1,3 +1,5 @@
+from typing import Optional
+
 import httpx
 
 from core.tools.errors import ToolProviderCredentialValidationError
@@ -32,7 +34,12 @@ class FeishuRequest:
         return res.get("tenant_access_token")
 
     def _send_request(
-        self, url: str, method: str = "post", require_token: bool = True, payload: dict = None, params: dict = None
+        self,
+        url: str,
+        method: str = "post",
+        require_token: bool = True,
+        payload: Optional[dict] = None,
+        params: Optional[dict] = None,
     ):
         headers = {
             "Content-Type": "application/json",

api/core/tools/utils/parser.py

@@ -3,6 +3,7 @@ import uuid
 from json import dumps as json_dumps
 from json import loads as json_loads
 from json.decoder import JSONDecodeError
+from typing import Optional
 
 from requests import get
 from yaml import YAMLError, safe_load
@@ -16,7 +17,7 @@ from core.tools.errors import ToolApiSchemaError, ToolNotSupportedError, ToolPro
 class ApiBasedToolSchemaParser:
     @staticmethod
     def parse_openapi_to_tool_bundle(
-        openapi: dict, extra_info: dict = None, warning: dict = None
+        openapi: dict, extra_info: Optional[dict], warning: Optional[dict]
     ) -> list[ApiToolBundle]:
         warning = warning if warning is not None else {}
         extra_info = extra_info if extra_info is not None else {}
@@ -174,7 +175,7 @@ class ApiBasedToolSchemaParser:
 
     @staticmethod
     def parse_openapi_yaml_to_tool_bundle(
-        yaml: str, extra_info: dict = None, warning: dict = None
+        yaml: str, extra_info: Optional[dict], warning: Optional[dict]
     ) -> list[ApiToolBundle]:
         """
         parse openapi yaml to tool bundle
@@ -191,7 +192,7 @@ class ApiBasedToolSchemaParser:
         return ApiBasedToolSchemaParser.parse_openapi_to_tool_bundle(openapi, extra_info=extra_info, warning=warning)
 
     @staticmethod
-    def parse_swagger_to_openapi(swagger: dict, extra_info: dict = None, warning: dict = None) -> dict:
+    def parse_swagger_to_openapi(swagger: dict, extra_info: Optional[dict], warning: Optional[dict]) -> dict:
         """
         parse swagger to openapi
 
@@ -253,7 +254,7 @@ class ApiBasedToolSchemaParser:
 
     @staticmethod
     def parse_openai_plugin_json_to_tool_bundle(
-        json: str, extra_info: dict = None, warning: dict = None
+        json: str, extra_info: Optional[dict], warning: Optional[dict]
     ) -> list[ApiToolBundle]:
         """
         parse openapi plugin yaml to tool bundle
@@ -287,7 +288,7 @@ class ApiBasedToolSchemaParser:
 
     @staticmethod
     def auto_parse_to_tool_bundle(
-        content: str, extra_info: dict = None, warning: dict = None
+        content: str, extra_info: Optional[dict], warning: Optional[dict]
     ) -> tuple[list[ApiToolBundle], str]:
         """
         auto parse to tool bundle