Optional Refresh Token Handling

[ahmtydn]

RequestFlag System - Type-Safe Request Configuration

Overview

The Vexana NetworkManager now supports a type-safe request configuration system using the RequestFlag enum, replacing magic strings in the extra map. This enhances code maintainability, IDE support, and prevents runtime errors caused by typos.

📌 Related Issue: #129 - Add Optional Refresh Token Handling to NetworkManager

RequestFlag Enum

The RequestFlag enum currently includes:

• RequestFlag.disableRefreshToken: Disables automatic refresh token handling for a specific request.

Use cases:

• Login/logout flows (prevent infinite loops)
• Public API calls without authentication
• One-off authentication flows

// Example: disable refresh token for the login endpoint
final result = await networkManager.send<LoginRequest, LoginResponse>(
  '/auth/login',
  parseModel: LoginResponse(),
  method: RequestType.POST,
  data: loginData,
  requestFlags: {RequestFlag.disableRefreshToken},
);

Updated API Usage

Both send and sendRequest methods now accept an optional requestFlags parameter:

// send method
tFuture<IResponseModel<R?, E?>> send<T extends INetworkModel<T>, R>(
  String path, {
  required T parseModel,
  required RequestType method,
  // ... other parameters
  Set<RequestFlag>? requestFlags,
});

// sendRequest method
Future<NetworkResult<R, E>> sendRequest<T extends INetworkModel<T>, R>(
  String path, {
  required T parseModel,
  required RequestType method,
  // ... other parameters
  Set<RequestFlag>? requestFlags,
});

Additional utility extensions are available via RequestFlagExtension.

Benefits

1. Type Safety: Compile-time checks prevent typos, IDE autocomplete support.
2. Extensibility: Easily add new flags, maintain backward compatibility.
3. Developer Experience: Self-documenting code, easy refactoring.
4. Maintainability: Eliminates scattered magic strings, centralizes flag definitions.

Future Enhancements

Potential additional flags:

• disableRetry: Disable automatic retry mechanism
• disableErrorDialog: Disable error dialog display
• disableCache: Disable caching for specific requests
• enableDebugLogging: Enable detailed logging per request

Best Practices

• Always use a Set<RequestFlag> for consistency.
• Document the reason for each flag use in comments.
• Avoid unnecessary flags to keep code clean.

// ✅ Good
requestFlags: {RequestFlag.disableRefreshToken}

// ❌ Avoid
requestFlags: [RequestFlag.disableRefreshToken].toSet()

[ahmtydn]

Description This PR improves refresh token handling by removing the global handleRefreshToken parameter and introducing a per-request disableRefreshToken flag:

The disableRefreshToken parameter can now be set for each request in the send and sendRequest methods. When provided, disableRefreshToken is automatically added to options.extra['disableRefreshToken'] internally, so users do not need to manage options.extra themselves. This change prevents side effects and race conditions that could occur when using a global parameter for concurrent requests.

Summary:

• Removed global handleRefreshToken.

• Added per-request disableRefreshToken parameter.

• The library handles setting options.extra['disableRefreshToken'] automatically.

• Safer and more predictable refresh token handling for concurrent requests.

[ahmtydn]

Pull Request Overview

This PR introduces an optional refresh token handling mechanism in the NetworkManager, allowing developers to toggle token refresh for individual requests.

• Adds a new nullable boolean parameter in multiple methods to control refresh token handling
• Propagates the new parameter via request options and integrates it with error interception
• Updates tests and interface definitions to support the new behavior

Reviewed Changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated no comments.

Show a summary per file Comments suppressed due to low confidence (2)

Insignificant comment. He seems confused by previous statements.

[ahmtydn]

Can you review the final version? @VB10