fix: 更新版本号至 0.7.0.beta1

2026-06-06 19:08:57 +08:00 · 2026-06-06 19:08:57 +08:00 · 2e8ef62c71
commit 2e8ef62c71
parent d19264bf77
12 changed files with 36 additions and 50 deletions
--- a/README.en.md
+++ b/README.en.md
@ -42,6 +42,9 @@ Yuxi is an LLM-powered platform for building knowledge-base and knowledge-graph
 | Doc parsing | MinerU · PaddleX · RapidOCR |
 | Deployment | Docker Compose |

+
+![image-20260606190609377](https://xerrors.oss-cn-shanghai.aliyuncs.com/github/image-20260606190609377.png)
+
 ## Quick Start

 **Prerequisites**: [Docker](https://docs.docker.com/get-docker/) and Docker Compose installed, plus at least one OpenAI-compatible LLM API.
@ -49,7 +52,7 @@ Yuxi is an LLM-powered platform for building knowledge-base and knowledge-graph
 **1. Clone and initialize**

 ```bash
-git clone --branch v0.7.0.dev1 --depth 1 https://github.com/xerrors/Yuxi.git
+git clone --branch v0.7.0.beta1 --depth 1 https://github.com/xerrors/Yuxi.git
 cd Yuxi

 # Linux/macOS
--- a/README.md
+++ b/README.md
@ -44,12 +44,17 @@
 | 文档解析 | MinerU · PaddleX · RapidOCR |
 | 部署 | Docker Compose |

+![image-20260606190609377](https://xerrors.oss-cn-shanghai.aliyuncs.com/github/image-20260606190609377.png)
+
 ## 最新动态


 <details open>
 <summary>[2026/06] v0.7.0 开发中（重要不兼容变更）</summary>

+
+</details>
+
 ### 重大变更

 - **模型配置收敛**：移除旧版 v1 模型配置与 Ollama 支持，运行时统一使用 `provider_id:model_id` 与独立 provider 模块，自定义 provider 迁移到数据库
@ -58,41 +63,13 @@
 - **Skill 安装与权限收敛**：以 `source_type / share_config / enabled` 表达来源、生效范围与启用状态；内置 Skill 启动自动入库并默认全局启用，上传/远程统一改为「解析草稿 → 确认安装」
 - **用户身份命名收敛**：业务登录标识统一为字符串 `uid`

+
+
 详见 [开发路线图](docs/develop-guides/roadmap.md)。
-
-</details>
-
 <details>
 <summary>[2026/04/01] v0.6.0 版本发布</summary>

-### 新增
-
- 重构后端代码 src -> backend/package/yuxi
- 重构文档解析，统一文档解析体验，并新增 Parser 类
- 新增 LITE 模式启动，启动时不加载知识库、知识图谱相关模块，可以使用 make up-lite 快捷启动
- 新增沙盒环境，详见后续文档更新，统一沙盒虚拟路径前缀默认值为 `/home/gem/user-data`
- 新增基于沙盒的文件系统，前端工作台可以查看文件系统，支持预览（文本、图片、PDF、HTML）、下载文件
- 新增 `present_artifacts` 内置工具：Agent 可将 `/home/gem/user-data/outputs/` 下的结果文件显式写入 LangGraph state 的 `artifacts` 字段，前端支持在输入框顶部以默认折叠的堆叠卡片展示本轮交付物文件，并保持可下载、可预览能力
- 新增基于沙盒的知识库只读映射，按“用户可访问知识库 ∩ 当前 Agent 已启用知识库”暴露原始文件与解析后的 Markdown
- 重构附件系统，直接集成在了沙盒文件系统中，附件上传后直接落盘到沙盒挂载目录
- 优化前端流式消息体验：新增通用 `useStreamSmoother` 调度层，统一平滑 Agent runs SSE、普通聊天流与审批恢复流中的 `loading` chunk
- 优化项目文档说明，并添加贡献指南
- 重构前端 Agent 路由结构，体验更加顺畅，切换更加自然（类 chatgpt 体验）
- 新增 API Key 认证功能，支持外部系统通过 API Key 调用系统服务
- 新增 subagents 的支持，支持在 web 中添加 subagents，以及两个内置的子智能体
- 新增内置Skills reporter，并移除内置 Agent reporter，数据库报表将由 Skills 完成
- 新增内置 Skills `deep-reporter`，用于指导生成科研报告、行业调研和其他深度分析类长报告
- 重构内置 Skills/MCP/Subagents 安装/添加/移除机制：内置 skill 支持按需安装、基于 `version + content_hash` 的更新提示与覆盖确认，不再使用服务器级开关切换
- 新增知识库 PDF、图片的预览功能
- 重构后端测试目录结构：按 `unit / integration / e2e` 分层迁移现有测试，拆分全局 `conftest.py`，统一测试入口为 `uv run --group test pytest`，并新增独立测试规范文档 `docs/develop-guides/testing-guidelines.md`
-
-
-### 修复
-
- 修复 LightRAG 知识库修改配置后，模型没有切换的 bug [#580](https://github.com/xerrors/Yuxi/issues/580)
- 修复数据库获取接口未过滤文件字段而导致的数据包过大的情况
- 修复 Thread 未绑定 agent_config_id 导致的历史对话切换后上下文配置错乱的问题
-
+详见 [changelog](docs/develop-guides/changelog.md)

 </details>

@ -124,7 +101,7 @@
 **1. 克隆代码并初始化**

 ```bash
-git clone --branch v0.7.0.dev1 --depth 1 https://github.com/xerrors/Yuxi.git
+git clone --branch v0.7.0.beta1 --depth 1 https://github.com/xerrors/Yuxi.git
 cd Yuxi

 # Linux/macOS
--- a/REFACTOR.md
+++ b/REFACTOR.md
@ -73,6 +73,7 @@
 - [x] 检查前端的这些 store 的使用，是否有需要调整优化收敛的地方
 - [x] 现在输入区域对于不同 mention 的渲染的 ICON 和 human-message 里面的渲染的 ICON;
 - [ ] 点开对话的时候要能够自动定位到尾部，而不是最开始。
+- [ ] office 组件预览，docx/pptx 可以转PDF，然后前端预览

 ## 子智能体设计

--- a/backend/package/pyproject.toml
+++ b/backend/package/pyproject.toml
@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"

 [project]
 name = "yuxi"
-version = "0.7.0.dev1"
+version = "0.7.0.beta1"
 description = "Yuxi 智能知识库与知识图谱平台核心后端包"
 readme = "README.md"
 requires-python = ">=3.12,<3.14"
--- a/backend/package/uv.lock
+++ b/backend/package/uv.lock
@ -5264,7 +5264,7 @@ wheels = [

 [[package]]
 name = "yuxi"
-version = "0.7.0.dev1"
+version = "0.7.0.beta1"
 source = { editable = "." }
 dependencies = [
    { name = "agent-sandbox" },
--- a/backend/pyproject.toml
+++ b/backend/pyproject.toml
@ -1,6 +1,6 @@
 [project]
 name = "yuxi-workspace"
-version = "0.7.0.dev1"
+version = "0.7.0.beta1"
 description = "基于大模型的智能知识库与知识图谱智能体开发平台，融合了 RAG 技术与知识图谱技术，基于 LangGraph v1 + Vue.js + FastAPI + LightRAG 架构构建"
 readme = "README.md"
 requires-python = ">=3.12,<3.14"
--- a/backend/uv.lock
+++ b/backend/uv.lock
@ -5489,7 +5489,7 @@ wheels = [

 [[package]]
 name = "yuxi"
-version = "0.7.0.dev1"
+version = "0.7.0.beta1"
 source = { editable = "package" }
 dependencies = [
    { name = "agent-sandbox" },
@ -5646,7 +5646,7 @@ requires-dist = [

 [[package]]
 name = "yuxi-workspace"
-version = "0.7.0.dev1"
+version = "0.7.0.beta1"
 source = { virtual = "." }
 dependencies = [
    { name = "arq" },
--- a/docker-compose.prod.yml
+++ b/docker-compose.prod.yml
@ -30,7 +30,7 @@ services:
    build:
      context: .
      dockerfile: docker/api.Dockerfile
-    image: yuxi-api:${YUXI_VERSION:-0.7.0.dev1}
+    image: yuxi-api:${YUXI_VERSION:-0.7.0.beta1}
    container_name: api-prod
    working_dir: /app
    networks:
@ -68,7 +68,7 @@ services:
    build:
      context: .
      dockerfile: docker/api.Dockerfile
-    image: yuxi-api:${YUXI_VERSION:-0.7.0.dev1}
+    image: yuxi-api:${YUXI_VERSION:-0.7.0.beta1}
    container_name: worker-prod
    working_dir: /app
    networks:
@ -145,7 +145,7 @@ services:
      context: .
      dockerfile: docker/web.Dockerfile
      target: production
-    image: yuxi-web:${YUXI_VERSION:-0.7.0.dev1}
+    image: yuxi-web:${YUXI_VERSION:-0.7.0.beta1}
    container_name: web-prod
    ports:
      - "80:80"
--- a/docker-compose.yml
+++ b/docker-compose.yml
@ -38,7 +38,7 @@ services:
    build:
      context: .
      dockerfile: docker/api.Dockerfile
-    image: yuxi-api:${YUXI_VERSION:-0.7.0.dev1}
+    image: yuxi-api:${YUXI_VERSION:-0.7.0.beta1}
    container_name: api-dev
    working_dir: /app
    volumes:
@ -86,7 +86,7 @@ services:
    build:
      context: .
      dockerfile: docker/api.Dockerfile
-    image: yuxi-api:${YUXI_VERSION:-0.7.0.dev1}
+    image: yuxi-api:${YUXI_VERSION:-0.7.0.beta1}
    container_name: worker-dev
    working_dir: /app
    volumes:
@ -125,7 +125,7 @@ services:
    build:
      context: ./docker/sandbox_provisioner
      dockerfile: Dockerfile
-    image: yuxi-sandbox-provisioner:${YUXI_VERSION:-0.7.0.dev1}
+    image: yuxi-sandbox-provisioner:${YUXI_VERSION:-0.7.0.beta1}
    container_name: sandbox-provisioner
    volumes:
      - ./docker/volumes/yuxi:/app/saves
@ -179,7 +179,7 @@ services:
      context: .
      dockerfile: docker/web.Dockerfile
      target: development
-    image: yuxi-web:${YUXI_VERSION:-0.7.0.dev1}
+    image: yuxi-web:${YUXI_VERSION:-0.7.0.beta1}
    container_name: web-dev
    volumes:
      - ./web/src:/app/src
--- a/docs/intro/quick-start.md
+++ b/docs/intro/quick-start.md
@ -20,7 +20,7 @@

 ```bash
 # 克隆最新版本
-git clone --branch v0.7.0.dev1 --depth 1 https://github.com/xerrors/Yuxi.git
+git clone --branch v0.7.0.beta1 --depth 1 https://github.com/xerrors/Yuxi.git
 cd Yuxi
 ```

--- a/scripts/bump-version.sh
+++ b/scripts/bump-version.sh
@ -65,6 +65,7 @@ echo "  - web/package.json"
 echo "  - docker-compose.yml"
 echo "  - docker-compose.prod.yml"
 echo "  - backend/uv.lock"
+echo "  - backend/package/uv.lock"
 if [ "$DEV_MODE" = false ]; then
    echo "  - README.md"
    echo "  - README.en.md"
@ -120,6 +121,10 @@ perl -0pi -e "s/(^name = \"yuxi\"\nversion = \")[^\"]+/\${1}${NEW_VERSION}/m" \
 perl -0pi -e "s/(^name = \"yuxi-workspace\"\nversion = \")[^\"]+/\${1}${NEW_VERSION}/m" \
    "${PROJECT_ROOT}/backend/uv.lock"

+echo "→ 更新 backend/package/uv.lock"
+perl -0pi -e "s/(^name = \"yuxi\"\nversion = \")[^\"]+/\${1}${NEW_VERSION}/m" \
+    "${PROJECT_ROOT}/backend/package/uv.lock"
+
 # -----------------------------------------------------------------------------
 # 6. 更新文档中的版本引用
 # -----------------------------------------------------------------------------
@ -127,15 +132,15 @@ perl -0pi -e "s/(^name = \"yuxi-workspace\"\nversion = \")[^\"]+/\${1}${NEW_VERS
 # 发布历史记录（如 [2026/04/01] v0.6.1 版本发布）不修改，保持为历史版本记录
 if [ "$DEV_MODE" = false ]; then
    echo "→ 更新 README.md"
-    perl -pi -e "s/(git clone --branch v)[0-9]+\\.[0-9]+\\.[0-9]+/\${1}${NEW_VERSION}/g" \
+    perl -pi -e "s/(git clone --branch v)[0-9]+\\.[0-9]+\\.[0-9]+(\\.[a-zA-Z0-9]+)?/\${1}${NEW_VERSION}/g" \
        "${PROJECT_ROOT}/README.md"

    echo "→ 更新 README.en.md"
-    perl -pi -e "s/(git clone --branch v)[0-9]+\\.[0-9]+\\.[0-9]+/\${1}${NEW_VERSION}/g" \
+    perl -pi -e "s/(git clone --branch v)[0-9]+\\.[0-9]+\\.[0-9]+(\\.[a-zA-Z0-9]+)?/\${1}${NEW_VERSION}/g" \
        "${PROJECT_ROOT}/README.en.md"

    echo "→ 更新 docs/intro/quick-start.md"
-    perl -pi -e "s/(git clone --branch v)[0-9]+\\.[0-9]+\\.[0-9]+/\${1}${NEW_VERSION}/g" \
+    perl -pi -e "s/(git clone --branch v)[0-9]+\\.[0-9]+\\.[0-9]+(\\.[a-zA-Z0-9]+)?/\${1}${NEW_VERSION}/g" \
        "${PROJECT_ROOT}/docs/intro/quick-start.md"
 else
    echo "→ dev 模式，跳过 README.md、README.en.md 和 docs/intro/quick-start.md 的分支版本更新"
--- a/web/package.json
+++ b/web/package.json
@ -1,6 +1,6 @@
 {
  "name": "yuxi-web",
-  "version": "0.7.0.dev1",
+  "version": "0.7.0.beta1",
  "private": true,
  "type": "module",
  "scripts": {