Skip to content

README.md

Latest commit

 

History

History
132 lines (86 loc) · 4.3 KB

README.md

File metadata and controls

132 lines (86 loc) · 4.3 KB

FeatureFlags

This example shows how to treat feature flags as first class state mutations with auditability, history inspection, and approval style restrictions.

It is good fit when you want to see how the engine behaves in an operational workflow where toggles may be simple, but the decision process around them is not.

Domain

The state is dictionary of feature names to enabled or disabled values.

The example covers three workflows:

  • enabling feature
  • disabling feature
  • toggling multiple features in batch

What this example demonstrates

  • mutation driven feature rollout
  • policy checks around sensitive toggles
  • batch execution with mixed outcomes
  • history retrieval and log printing
  • using MutationContext metadata for approval tracking

Project structure

How it works

Program.cs:

  1. registers the engine with strict options
  2. resolves IMutationEngine
  3. registers RequireTwoManApprovalPolicy
  4. runs the example scenarios
  5. prints history for the main state
  6. prints engine statistics

The sample also includes BusinessHoursPolicy as reference implementation. It is not registered by default, but it shows how time based operational limits would be expressed.

Mutation flow

Enable feature

EnableFeatureMutation enables a flag.

  • validates that the feature name exists
  • writes the new enabled value into copied state
  • emits ChangeSet entry for the feature

Disable feature

DisableFeatureMutation disables a flag.

  • validates that the feature exists
  • writes the new disabled value into copied state
  • emits change record for the flag

Policies

Two man approval

RequireTwoManApprovalPolicy blocks certain high risk feature changes unless metadata contains two distinct approvers.

It demonstrates:

  • inspection of MutationContext.Metadata
  • policy decisions based on the mutation type
  • excluding the actor from the approval list

Business hours

BusinessHoursPolicy is simple time based policy for highrisk changes.

Use it as reference if you want to restrict rollout windows.

Scenarios

Enable new checkout

EnableNewCheckoutScenario shows how new feature is turned on from known starting state.

Disable legacy checkout

DisableLegacyCheckoutScenario shows how feature can be turned off with approval metadata attached.

Batch feature toggle

BatchFeatureToggleScenario shows mixed feature operations in single batch.

It demonstrates:

  • batch mutation submission
  • partial success and failure handling
  • result logging
  • folding the batch output back into the final state

What to read first

  1. Program.cs
  2. State/FeatureFlagsState.cs
  3. Mutations/EnableFeatureMutation.cs
  4. Mutations/DisableFeatureMutation.cs
  5. Policies/RequireTwoManApprovalPolicy.cs
  6. Scenarios/BatchFeatureToggleScenario.cs

Run

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

Expected output

You should see:

  • one or more feature flag changes
  • batch mutation logs
  • a history dump for the tracked state
  • a statistics summary at the end

The sample is intentionally written as an operational console app, not as presentation layer.