This plugin system provides a flexible and extensible architecture for defining and executing plugins. It supports named inputs and outputs, node-based workflows, and dynamic execution of specific functions. Designed for advanced use cases like FX graphs or modular applications, this system enables developers to build robust, reusable, and dynamic components.
- Strongly-Typed Inputs and Outputs: Each plugin can define named inputs and outputs with automatic mapping.
- Dynamic Function Registration: Register multiple output functions for plugins and execute them only when connected.
- Node-Based Workflow: Create and connect plugins dynamically in a graph-style system.
- Single Plugin Execution: Execute individual plugins with specified inputs and outputs.
- Last Result Propagation: Pass the output of one plugin to the next in sequence.
- Save and Load Graphs: Serialize and deserialize the plugin graph for reuse.
- UI-Friendly Properties: Includes positional and connection metadata for rendering plugins in a graphical node editor.
- Clone the repository or download the source files.
- Add the files to your project.
Create a custom plugin by subclassing Plugin
:
using PluginSystem;
public class ConsumerPlugin : Plugin
{
public override string PluginId => "consumer_plugin";
public override string PluginName => "Consumer Plugin";
public override FXPluginType PluginType => FXPluginType.Sequential;
public override List<string> InputNames => new List<string> { "Input" };
public override List<string> OutputNames => new List<string> { "ProcessedOutput" };
public ConsumerPlugin()
{
RegisterFunction("ProcessedOutput", (inputs, lastResult) =>
{
var input = Convert.ToSingle(inputs["Input"]);
return input * 2; // Example processing
});
}
}
Run a single plugin by creating an instance and setting its inputs:
var mathPlugin = new MathPlugin();
var inputs = new Dictionary<string, object>
{
{ "A", 10f },
{ "B", 5f }
};
var sumResult = mathPlugin.ExecuteFunction("Sum", inputs, null);
Console.WriteLine($"Sum: {sumResult}");
var differenceResult = mathPlugin.ExecuteFunction("Difference", inputs, null);
Console.WriteLine($"Difference: {differenceResult}");
- Create an
FXGraph
and add plugins to it. - Connect plugins dynamically by specifying source and target nodes.
- Run the graph to execute connected functions.
Example:
static async Task Main()
{
var fxGraph = new FXGraph();
// Create the MathPlugin
var mathPlugin = new MathPlugin();
// Add MathPlugin to the graph
var mathNode = new FXGraphItem(mathPlugin);
fxGraph.AddPlugin(mathPlugin);
// Simulate an external consumer node (hypothetical)
var consumerPlugin = new ConsumerPlugin(); // A plugin that consumes "Sum"
fxGraph.AddPlugin(consumerPlugin);
// Connect the "Sum" output of MathPlugin to the "Input" of ConsumerPlugin
fxGraph.ConnectNodes(mathNode.UniqueId, "Sum", consumerPlugin.PluginId, "Input");
// Define inputs for the MathPlugin
var inputs = new Dictionary<string, object> { { "A", 10f }, { "B", 5f } };
// Run the graph
var result = await fxGraph.RunFXGraphAsync();
Console.WriteLine($"Final Result: {result}");
}
Just incase you don't want your UI to freeze while executing a plugin, you can run the graph asynchronously.
var mathPlugin = new MathPlugin();
var inputs = new Dictionary<string, object>
{
{ "A", 10f },
{ "B", 5f }
};
var sumResult = await mathPlugin.ExecuteFunctionAsync("Sum", inputs, null);
Console.WriteLine($"Sum: {sumResult}");
fxGraph.SaveToFile("fxgraph.json");
var loadedGraph = FXGraph.LoadFromFile("fxgraph.json", availablePlugins);
Contributions are welcome! Feel free to submit issues, fork the repo, or create pull requests.
This project is licensed under the Apache 2.0 License. See LICENSE
for details.
For questions or support, fill out an issue.