How to Build from Source#

This document provides step-by-step instructions for building the ChatQnA Core sample application and File Watcher service from source. Refer to the prerequisites section in the guide to install the appropriate software dependencies.

Build ChatQnA Core from Source#

For detailed instructions on building from source, visit the Build from Source Guide.

ChatQnA Core Docker Compose Deployment#

For docker compose deployment instructions, visit the Running Application Container Guide.

Build File Watcher Service from Source#

In the Windows® environment, the File Watcher Service works with the HMI application to continuously monitor file system activities such as creation, modification, and deletion. When it detects any changes, it sends the relevant file data over the network to the ChatQnA Core service for ingestion and contextual processing, supporting Retrieval-Augmented Generation (RAG) workflows.

Prerequisites#

Python Installer: Visit the official Python website. Select the latest version available under the “Python Releases for Windows” section.
Git[OPTIONAL]: Visit the official GIT website to download the executable

Build File Watcher Service in Windows#

To build the File Watcher executable binary, follow these steps:

Clone and download the source code by either using Git clone or downloading the source code as a ZIP file directly from the repository.
```
git clone https://github.com/open-edge-platform/edge-ai-suites.git edge-ai-suites
```
Set up the Virtual Environment with Python venv.
- Open Command Prompt.
  - Press Win + R, type cmd, and hit Enter.
- Create Virtual Environment.
```
# Replace `<venv_name>` with your preferred name.
python -m venv <venv_name>
```
Activate the Virtual Environment.
```
<venv_name>\Scripts\activate
```
Once it’s activated, the environment name appears in parentheses as follows:
```
(venv_name) C:\Users\YourName\project>
```

Navigate to the Project folder downloaded in Step 1.

cd edge-ai-suites\manufacturing-ai-suite\hmi-augmented-worker\file_watcher

Install Packages Inside the Virtual Environment.

pip install -r requirements.txt --no-cache-dir

If your system is behind a proxy, do as follows:

# Replace <your_proxy> and <port> to your network proxy and port number
pip install -r requirements.txt --no-cache-dir --proxy <your_proxy>:<port>

On Windows®, typically, proxy information can be fetched using the command,

netsh winhttp show proxy

This will output one of the following two outputs:

Current WinHTTP proxy settings:
     Direct access (no proxy server).

or

Current WinHTTP proxy settings:
     Proxy Server:  <your_proxy>:<port>
     Bypass List:   <bypass_list>

Set up Environment Variables using .bat.

To configure the file watcher service, you need to set up the environment variables using the set_env_vars.bat file provided. Follow the steps below to ensure proper configuration:
- Open and edit the values for the variables with your corresponding setup.
  
  Open the set_env_vars.bat file and update the values for the environment variables according to your setup. Below are the detailed explanation:
  - WATCH_DIRECTORY:
    
    This should point to the directory you want the file watcher service to monitor. It should be a valid path on your system where files are expected to be added, monitored, or deleted. The service will track these events and send the file data to the backend for embedding creation.
    
    Important: Ensure that the directory specified in WATCH_DIRECTORY does not have access or permission restrictions. The file watcher service must have read and write permissions to monitor and process files within this directory effectively. And also only .txt, .pdf and .docx file format are only supported formats by backend for embedding creation.
    
    Examples of Valid Directory Paths:
    - On Windows:
      - C:\Users\YourUsername\Documents\WatchedFolder
      - D:\Data\Projects\MonitorFolder
  - DOCUMENT_ENDPOINT:
    
    This variable should be set to the URL of your backend service where the file data will be sent. Replace with the actual IP address or hostname of your backend service.
  - no_proxy:
    
    This variable is used to specify IP addresses or hostnames that should bypass the proxy settings if your system is behind proxy. Set this to the IP address of your backend service to ensure direct communication.
- Then, execute the .bat file as shown:
```
.\set_env_vars.bat
```
- After executing the set_env_vars.bat file, you should see no output if the scripts run successfully. The environment variables will be set in the current command prompt session, and you can verify them by running the following command:
```
echo %WATCH_DIRECTORY%
echo %DOCUMENT_ENDPOINT%
echo %no_proxy%
```
  These commands will display the values you have set for each variable, confirming that they are correctly configured.
Compile and build the File Watcher Service executable.
```
pyinstaller file_watcher.py -F --onefile
```
Execute the File Watcher Service executable.
- Before starting the File Watcher Service, ensure that your backend ChatQnA Core service is up.
```
.\dist\file_watcher.exe
```
After the service starts, the file watcher continuously monitors file events occurring in the designated folder specified by WATCH_DIRECTORY in the set_env_vars.bat file, until it is stopped by a keyboard interrupt.

Troubleshooting#

`Error: [WinError 206] The filename or extension is too long`#

Description:#

When attempting to install packages using pip on Windows, you may encounter the following error:

ERROR: Could not install packages due to an OSError: [WinError 206] The filename or extension is too long: "...

Solution:#

Modify the registry entry using Registry Editor.
1. Open Registry Editor:
  - Press Win + R to open the Run dialog.
  - Type regedit and press Enter.
2. Navigate to the FileSystem Key:
  - Go to: HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\FileSystem
3. Modify LongPathsEnabled:
  - Find the LongPathsEnabled entry.
  - Double-click on it and set its value to 1.
  - Click OK to save the changes.

After enabling long path support, you may need to restart your computer for the changes to take effect. Once done, try running your pip install command again.

If the issue persists, consider using a shorter path for your project or virtual environment, as mentioned earlier. Moving your project to a directory with a shorter path can help avoid hitting the path length limit.