Plugin Management

Developing Plugins

Developers can develop plugins by referring to the Plugin Development Guide.

The plugin package must meet the following requirements:

1. The plugin package must be a ZIP package.
2. The plugin package must contain the following files:
   • config.json configuration file
   • Frontend product: The file must be a ZIP package, which contains the frontend asset directory and its files, and the index.html file.
   • Backend product: The file must be a ZIP package, which contains the dynamic libraries required by the plugin and a single dynamic library file of the corresponding platform and architecture. The key value of the backend product in the config.json configuration file is backend_{platform}_{machine}, where platform indicates the platform name and machine indicates the architecture name. For example, the key value of the backend product in the Linux x86 environment is backend_linux_x86_64.

   The config.json configuration file must meet the following requirements:

   json
   {
   "pluginName": "Plugin Name",
   "frontend": "Frontend product name", // ZIP package
   "backend_{platform}_{machine}": "Backend product name", // ZIP package or dynamic library
   }

   In the command, platform indicates the platform name, and machine indicates the architecture name.

3. The number of files in the plugin package cannot exceed 1000, and the size of a single file cannot exceed 200 MB.
4. The plugin package must be owned by the current user and have the read and write permissions. Link files and files containing links are not supported.

MindStudio Insight supports the loading of any plugin in the .so format. You must verify the integrity of the required plugin package to ensure that the package is from a secure and reliable source, thereby avoiding potential security risks such as community poisoning and malicious code injection.

Installing Plugins

Go to the installation directory of MindStudio Insight and run the following command to install the developed plugin package. plugin package path indicates the path of the plugin package.

python resources/profiler/plugin_install.py install --path=plugin package path

Using Plugins

After the installation is complete, start MindStudio Insight and import data.

If the wakeup logic is developed independently, use the plugin package based on the actual situation.