Diagnostics

PipeForge includes built-in diagnostics via the System.Diagnostics.DiagnosticSource API. This allows you to observe pipeline execution without modifying your steps.

Emitted Events

The pipeline emits the following diagnostic events for each step:

Event Name	Description
PipelineStep.Start	Emitted before a step is invoked. Payload includes the step.
PipelineStep.Stop	Emitted after a step completes. Payload includes the step.

All events are emitted under a listener name in the format:

PipeForge.PipelineRunner<TContext>

You can subscribe to these events to monitor or trace execution flow.

Example: Listening for Events

Here's an example of a diagnostic listener that logs every start and stop event:

using System.Diagnostics;

var listener = new DiagnosticListener("PipeForge.PipelineRunner<SampleContext>");
listener.Subscribe(new Observer((name, payload) =>
{
    Console.WriteLine($"{name}: {payload}");
}));

Integration Tips

• You can use this mechanism to generate timing metrics, debug issues, or visualize step execution.
• Diagnostic events are low-overhead and safe to leave enabled in production.
• Combine this with the Describe() method for a full picture of pipeline structure and execution behavior.

Conclusion

PipeForge diagnostics give you deep visibility into pipeline execution with minimal effort. Whether you're debugging a failing step or building runtime instrumentation, the diagnostics hooks are ready to help.