Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
50 changes: 50 additions & 0 deletions FAQ.md
Original file line number Diff line number Diff line change
Expand Up @@ -51,6 +51,23 @@ This is usually because the system lacks necessary runtime components.
It is not recommended to build the client locally by yourself, as you may encounter configuration or environmental issues.
It is recommended to download the official stable installation package (e.g., v1.1.2+) directly from the [Release page](https://github.com/iflytek/astron-rpa/releases).

### Q: 🆕 Can RPA be deployed and run in an offline intranet environment? How to import images?

**A:** ✅ **Yes!**
In an offline environment, you need to export image packages using `docker save` in an environment with internet access, and import them using `docker load` in the intranet. After importing, please use `docker images` to check if the image Tags are complete. If there are dangling images without names, please use `docker tag` to re-tag them.

### Q: 🆕 Installation of dependencies like `pywinhook` and `psycopg2` fails with errors when building the client locally (`build.bat`)?

**A:**
1. **C++ Build Environment**: Ensure Microsoft Visual C++ 14.0 or higher is installed (including MSVC v143 and Win10/11 SDK).
2. **Permission Issue**: Please run the build script with Administrator privileges.
3. **Dependency Trimming**: If specific database drivers (like Oracle/PostgreSQL) are not needed in the current scenario, you can modify `/engine/components/astronverse-database/pyproject.toml` and remove dependencies like `psycopg2` and `cx-oracle` to bypass complex local compilation.
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The dependency names mentioned here do not exactly match the ones defined in engine/components/astronverse-database/pyproject.toml. In that file, they are listed as psycopg2-binary and cx_Oracle. Using the exact names will help users find and remove them more easily, especially since cx_Oracle uses an underscore and specific casing.

Suggested change
3. **Dependency Trimming**: If specific database drivers (like Oracle/PostgreSQL) are not needed in the current scenario, you can modify `/engine/components/astronverse-database/pyproject.toml` and remove dependencies like `psycopg2` and `cx-oracle` to bypass complex local compilation.
3. **Dependency Trimming**: If specific database drivers (like Oracle/PostgreSQL) are not needed in the current scenario, you can modify `/engine/components/astronverse-database/pyproject.toml` and remove dependencies like `psycopg2-binary` and `cx_Oracle` to bypass complex local compilation.


### Q: 🆕 Does the open-source version of RPA support front-end and back-end separate deployment for secondary development and debugging?

**A:** ✅ **Yes!**
The front-end and back-end separate mode is fully supported and recommended for development and debugging scenarios. You can independently run and debug front-end and back-end code. Please make sure to pull the latest code for configuration.

---

## 👥 Client Related
Expand Down Expand Up @@ -340,6 +357,31 @@ http://{IP_ADDRESS}:32742/api/rpa-openapi/workflows/get
- **Input parameters:** Define "Process Parameters" in the RPA workflow design, and pass the corresponding Key-Value when calling externally.
- **Output parameters:** Return JSON data through HTTP request nodes or Python scripts, which can be referenced by subsequent nodes via variables.

### Q: 🆕 How does the main workflow pass parameters to a component?

**A:**

1. Set "Input Parameters" (or component properties) in the component.
2. Publish the component.
3. Drag the component into the main workflow again, and you can assign values to these parameters in the right configuration panel.
> ⚠️ **Note:** If the component's parameters are modified, it must be published first, and re-dragged into the main workflow to use the new parameters.

### Q: 🆕 Are component element names unique within a component? Do duplicate names across different components cause issues?

**A:** Element names are unique within a single component. Element names across different components can be duplicated without affecting each other.

### Q: 🆕 Does RPA web element positioning (XPath) support using variables?

**A:** ✅ **Yes!** You can use variables or parameters in the XPath path string to achieve dynamic element positioning.

### Q: 🆕 What is the default positioning method used for capturing web elements? What if the absolute path changes due to changes in the page structure?

**A:**

- When capturing an element, if the element has an ID, the ID will be locked by default (as part of the XPath).
- If the absolute path changes due to a web page structure change, but the ID remains the same, you can adapt by adjusting the ID's checkbox state.
- If the ID changes or is unstable, consider unchecking the ID attribute, and use other stable relative paths or attributes for positioning, or observe the page change patterns to manually modify the XPath.

### Q: What if dependency package downloads fail?

**A:**
Expand Down Expand Up @@ -416,6 +458,14 @@ docker logs [container_name] > logs.txt

It may be missing `Microsoft Edge WebView2 Runtime` or the version is too low (common in older systems or cloud desktops). Please try updating WebView2 Runtime.

### Q: 🆕 What should I do if the email component reports errors (e.g., search parameter mismatch, unknown encoding utf-8) when using imap4 to receive emails?

**A:** This is due to compatibility issues with specific email server interactions and encoding processing. It is recommended to update to the latest version of the client. The underlying component has fixed the encoding parsing and attachment name regex extraction null pointer issues. If encountered in older versions, you can contact technical support for a source code patch.

### Q: 🆕 Why does Excel automation abnormally open a new blank workbook when performing operations like "Copy Cells"?

**A:** This may be an instance conflict caused by the underlying scheduling mechanism. In the latest version, `application.py` has been optimized to prioritize taking over existing Excel instances and disable warning pop-ups by default. Please ensure you upgrade to the latest client version.

### Q: 🆕 Web element capturing feature frequently fails or causes the browser to crash?

**A:**
Expand Down
50 changes: 50 additions & 0 deletions FAQ.zh.md
Original file line number Diff line number Diff line change
Expand Up @@ -51,6 +51,23 @@
不建议在本地自行构建客户端,可能会遇到配置或环境问题。
建议直接从 [Release 页面](https://github.com/iflytek/astron-rpa/releases) 下载官方提供的稳定版安装包(如 v1.1.2+)。

### Q: 🆕 离线内网环境下可以部署和运行 RPA 吗?如何导入镜像?

**A:** ✅ **可以!**
在离线环境下,需要在有网环境使用 `docker save` 导出镜像包,并在内网使用 `docker load` 导入。导入后请使用 `docker images` 检查镜像 Tag 是否完整,如果出现无名称(dangling)镜像,请使用 `docker tag` 重新打标签。

### Q: 🆕 本地构建客户端 (`build.bat`) 时,安装 `pywinhook`、`psycopg2` 等依赖报错失败?

**A:**
1. **C++ 编译环境**:确保已安装 Microsoft Visual C++ 14.0 或更高版本(包含 MSVC v143 和 Win10/11 SDK)。
2. **权限问题**:请使用管理员权限(Administrator)运行构建脚本。
3. **依赖裁剪**:如果当前场景不需要特定数据库驱动(如 Oracle/PostgreSQL),可以修改 `/engine/components/astronverse-database/pyproject.toml`,移除 `psycopg2` 和 `cx-oracle` 等依赖,以绕过复杂的本地编译。
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The dependency names mentioned here do not exactly match the ones defined in engine/components/astronverse-database/pyproject.toml. In that file, they are listed as psycopg2-binary and cx_Oracle. Using the exact names will help users find and remove them more easily, especially since cx_Oracle uses an underscore and specific casing.

Suggested change
3. **依赖裁剪**:如果当前场景不需要特定数据库驱动(如 Oracle/PostgreSQL),可以修改 `/engine/components/astronverse-database/pyproject.toml`,移除 `psycopg2``cx-oracle` 等依赖,以绕过复杂的本地编译。
3. **依赖裁剪**:如果当前场景不需要特定数据库驱动(如 Oracle/PostgreSQL),可以修改 `/engine/components/astronverse-database/pyproject.toml`,移除 `psycopg2-binary``cx_Oracle` 等依赖,以绕过复杂的本地编译。


### Q: 🆕 开源版 RPA 支持前后端分离部署以便于二次开发调试吗?

**A:** ✅ **支持!**
前后端分离模式在开发调试场景下是完全支持且推荐的,可以独立运行和调试前端与后端代码。请确保拉取最新代码进行配置。

---

## 👥 客户端相关
Expand Down Expand Up @@ -340,6 +357,31 @@ http://{IP_ADDRESS}:32742/api/rpa-openapi/workflows/get
- **入参:** 在 RPA 流程设计中定义“流程参数”,外部调用时传入对应 Key-Value。
- **出参:** 通过 HTTP 请求节点或 Python 脚本返回 JSON 数据,后续节点可通过变量引用。

### Q: 🆕 主流程如何向组件传递参数?

**A:**

1. 在组件中设置“输入参数”(或组件属性)。
2. 发布该组件。
3. 在主流程中重新拖入该组件,即可在右侧配置面板中为这些参数赋值。
> ⚠️ **注意:** 如果组件修改了参数,必须先发版,并在主流程中重新拖入才能生效使用新参数。

### Q: 🆕 组件元素名称在组件内是唯一的吗?组件之间有重复名称影响吗?

**A:** 元素名称在单个组件内是唯一的。不同组件之间的元素名称可以重复,互不影响。

### Q: 🆕 RPA 的网页元素定位(XPath)支持使用变量吗?

**A:** ✅ **支持!** 可以在 XPath 的路径字符串中使用变量或参数,以实现动态元素定位。

### Q: 🆕 网页元素拾取默认使用什么定位方式?如果页面结构变化导致绝对路径改变怎么办?

**A:**

- 拾取元素时,如果元素有 ID,默认会锁定 ID(作为 XPath 的一部分)。
- 如果网页结构变动导致绝对路径变化,但 ID 没变,可以通过调整 ID 的勾选状态来适应。
- 如果 ID 变动或不稳定,可以考虑取消勾选 ID 属性,改用其他稳定的相对路径或属性进行定位,或者观察页面变动规律手动修改 XPath。

### Q: 依赖包下载失败怎么办?

**A:**
Expand Down Expand Up @@ -416,6 +458,14 @@ docker logs [container_name] > logs.txt

可能是缺少 `Microsoft Edge WebView2 Runtime` 或版本过低(常见于老旧系统或云桌面)。请尝试更新 WebView2 Runtime。

### Q: 🆕 邮件组件使用 imap4 接收邮件时报错(如 search 参数不匹配、unknown encoding utf-8 等)怎么办?

**A:** 这是由于特定邮件服务器交互和编码处理的兼容性问题。建议更新到最新版本的客户端,底层组件已修复编码解析与附件名正则提取空指针问题。在旧版本中遇到时可联系技术支持获取源码补丁。

### Q: 🆕 Excel 自动化执行“复制单元格”等操作时,为什么会异常自动打开新的空白工作簿?

**A:** 可能是底层调度机制导致的实例冲突。在最新的版本中,已优化了 `application.py`,优先接管已存在的 Excel 实例并默认禁用警告弹窗。请确保升级到最新版客户端。

### Q: 🆕 网页元素拾取功能频频失效或导致浏览器闪退?

**A:**
Expand Down
Loading