Skip to content

README.md

Latest commit

 

History

History
166 lines (111 loc) · 5.36 KB

README.md

File metadata and controls

166 lines (111 loc) · 5.36 KB

WorkflowApprovals

This example models multi step workflow where ordering matters and approvals can be blocked by policy.

It is the best sample to read if you want to understand how the engine behaves when one mutation must happen before the next one can safely proceed.

Domain

The domain state contains:

  • a workflow ID
  • an initiator
  • an ordered list of workflow steps

Each step has status and may be approved or rejected by user.

The example includes two scenarios:

  • a happy path where the steps are approved in sequence
  • a rejection path where one or more approvals are blocked
  • a side effects path where workflow start and rejection emit observable side effects

What this example demonstrates

  • ordered workflow mutation
  • approval and rejection transitions
  • policy checks that depend on prior state
  • mutation context usage for audit and traceability
  • failure handling when step is out of order or unauthorized
  • side effect emission for monitoring, alerting, and follow-up workflows

Project structure

How it works

Program.cs:

  1. registers the engine with strict options
  2. resolves IMutationEngine
  3. registers workflow policies
  4. runs both scenarios
  5. prints engine statistics

The sample is intentionally sequential. It shows how stateful process can be advanced step by step while the engine keeps the mutation logic and policy logic separate.

Mutation flow

Start workflow

StartApprovalMutation creates workflow from list of step names.

  • validates initiator and step list
  • creates workflow steps from names
  • initializes a new workflow ID
  • emits change entry for the created step list
  • emits a WorkflowStarted side effect through SideEffect.Create(...)

Approve step

ApproveStepMutation approves a single workflow step.

  • validates the step index
  • validates approver input
  • updates the targeted step
  • emits change entry for the step status

Reject workflow

RejectWorkflowMutation marks the whole workflow as rejected.

  • applies rejection to every step
  • records the actor who rejected the workflow
  • emits workflow level change
  • emits a critical WorkflowRejected side effect through SideEffect.Critical(...)

Policies

Enforce order

EnforceOrderPolicy prevents a step from being approved before the previous step has been approved.

This is the main rule that makes the example a proper ordered workflow.

Require manager approval

RequireManagerApprovalPolicy limits approvals to designated managers.

This policy demonstrates:

  • checking the specific mutation type
  • validating approver identity
  • rejecting unauthorized actors

Scenarios

Happy path

HappyPathScenario starts workflow and approves each step in order.

It shows:

  • successful workflow initialization
  • sequential approvals
  • final state printing

Rejected path

RejectedScenario starts workflow and then attempts approvals that may be blocked.

It shows:

  • policy driven rejection
  • per step logging
  • final workflow state inspection

Side effects path

SideEffectsScenario starts a workflow and then rejects it to show side effects in the result object.

It shows:

  • a standard side effect created with SideEffect.Create(...)
  • a critical side effect created with SideEffect.Critical(...)
  • how Severity, RequiresAction, and Data can be read from MutationResult.SideEffects

What to read first

  1. State/ApprovalWorkflowState.cs
  2. State/WorkflowStep.cs
  3. State/StepStatus.cs
  4. Mutations/StartApprovalMutation.cs
  5. Mutations/ApproveStepMutation.cs
  6. Policies/EnforceOrderPolicy.cs
  7. Policies/RequireManagerApprovalPolicy.cs
  8. Scenarios/SideEffectsScenario.cs

Run

dotnet run --project Examples/Core/WorkflowApprovals/WorkflowApprovals.csproj

Expected output

You should see:

  • workflow start output
  • approvals or policy rejections for individual steps
  • the final workflow state
  • engine statistics at the end

This sample is the cleanest one to study for staged business processes and ordered execution rules.