mobile-networking

name	mobile-networking
description	Design and implement mobile networking layers — HTTP clients, interceptors, authentication token management, caching, offline queues, GraphQL, WebSockets, and response parsing across Flutter (Dio/http), Android (Retrofit/OkHttp/Ktor), and iOS (URLSession/Alamofire). TRIGGER when: user says /mobile-networking, asks about API integration in mobile, needs to set up networking, implement token refresh, or optimize API calls.
argument-hint	[networking task or API integration need]
user-invocable	true

Mobile Networking

You are a senior mobile engineer. Help the user design, implement, or review mobile networking layers with platform-specific best practices.

Process

Step 1: Understand Requirements

Question	Why It Matters
What API protocol? (REST, GraphQL, gRPC, WebSocket)	Determines client library
What auth mechanism? (Bearer JWT, OAuth2, API key, session cookie)	Token management strategy
Does the app work offline?	Offline queue and cache strategy
What is the expected request volume?	Connection pooling, batching
Are there file uploads/downloads?	Multipart, streaming, progress tracking
What serialization format? (JSON, Protocol Buffers, MessagePack)	Parser selection

Step 2: Choose HTTP Client

Flutter

Library	Best For	Features
Dio	Most apps — recommended default	Interceptors, cancellation, multipart, FormData, transformers
http	Simple apps, minimal dependencies	Lightweight, built by Dart team
Chopper	Retrofit-style code generation	Type-safe, interceptors, converters
graphql_flutter	GraphQL APIs	Queries, mutations, subscriptions, cache

// Dio — setup with interceptors
final dio = Dio(BaseOptions(
  baseUrl: 'https://api.example.com/v1',
  connectTimeout: const Duration(seconds: 10),
  receiveTimeout: const Duration(seconds: 15),
  headers: {'Content-Type': 'application/json'},
));

// Add interceptors
dio.interceptors.addAll([
  AuthInterceptor(tokenStorage),
  LogInterceptor(requestBody: true, responseBody: true),
  RetryInterceptor(dio: dio, retries: 3),
]);

// Type-safe API call
Future<List<Product>> getProducts() async {
  final response = await dio.get('/products');
  return (response.data as List)
      .map((json) => Product.fromJson(json))
      .toList();
}

// Multipart file upload
Future<void> uploadImage(File file) async {
  final formData = FormData.fromMap({
    'image': await MultipartFile.fromFile(file.path),
    'description': 'Product image',
  });
  await dio.post('/upload', data: formData);
}

Android (Kotlin)

Library	Best For	Features
Retrofit + OkHttp	Most apps — recommended default	Type-safe, interceptors, converters, widely adopted
Ktor Client	KMP / coroutines-first	Multiplatform, DSL-based, engine-agnostic
Apollo Kotlin	GraphQL APIs	Queries, mutations, subscriptions, normalized cache
OkHttp (raw)	Low-level control	Full HTTP control, WebSocket built-in

// Retrofit + OkHttp — setup
interface ProductApi {
    @GET("products")
    suspend fun getProducts(): List<ProductDto>

    @GET("products/{id}")
    suspend fun getProduct(@Path("id") id: String): ProductDto

    @POST("products")
    suspend fun createProduct(@Body product: CreateProductRequest): ProductDto

    @Multipart
    @POST("upload")
    suspend fun uploadImage(@Part image: MultipartBody.Part): UploadResponse
}

// OkHttp client with interceptors
val okHttpClient = OkHttpClient.Builder()
    .addInterceptor(AuthInterceptor(tokenStorage))
    .addInterceptor(HttpLoggingInterceptor().apply {
        level = if (BuildConfig.DEBUG) Level.BODY else Level.NONE
    })
    .connectTimeout(10, TimeUnit.SECONDS)
    .readTimeout(15, TimeUnit.SECONDS)
    .retryOnConnectionFailure(true)
    .build()

// Retrofit instance
val retrofit = Retrofit.Builder()
    .baseUrl("https://api.example.com/v1/")
    .client(okHttpClient)
    .addConverterFactory(MoshiConverterFactory.create(moshi))
    
    .build()

 api = retrofit.create(ProductApi::.java)

Android (Java)

// Retrofit + OkHttp — Java
public interface ProductApi {
    @GET("products")
    Call<List<ProductDto>> getProducts();

    @GET("products/{id}")
    Call<ProductDto> getProduct(@Path("id") String id);

    @POST("products")
    Call<ProductDto> createProduct(@Body CreateProductRequest product);
}

// With RxJava
public interface ProductApi {
    @GET("products")
    Single<List<ProductDto>> getProducts();
}

// OkHttp client
OkHttpClient client = new OkHttpClient.Builder()
    .addInterceptor(new AuthInterceptor(tokenStorage))
    .addInterceptor(new HttpLoggingInterceptor().setLevel(Level.BODY))
    .connectTimeout(10, TimeUnit.SECONDS)
    .build();

Retrofit retrofit = new Retrofit.Builder()
    .baseUrl("https://api.example.com/v1/")
    .client(client)
    .addConverterFactory(GsonConverterFactory.create())
    .addCallAdapterFactory(RxJava3CallAdapterFactory.create()) // for RxJava
    .build();

iOS (Swift)

Library	Best For	Features
URLSession + async/await	Most apps — recommended default	Built-in, no dependencies, full control
Alamofire	Apps needing request chaining, interceptors, retry	Battle-tested, rich feature set
Apollo iOS	GraphQL APIs	Queries, mutations, subscriptions, code generation
URLSession (Combine)	Reactive chains	dataTaskPublisher, built-in

// URLSession + async/await — modern approach
class APIClient {
    private let session: URLSession
    private let baseURL: URL
    private let decoder = JSONDecoder()

    func getProducts() async throws -> [Product] {
        let url = baseURL.appendingPathComponent("products")
        var request = URLRequest(url: url)
        request.setValue("Bearer \(token)", forHTTPHeaderField: "Authorization")

        let (data, response) = try await session.data(for: request)
        guard let httpResponse = response as? HTTPURLResponse,
              200..<300 ~= httpResponse.statusCode else {
            throw APIError.invalidResponse
        }
        return try decoder.decode([Product].self, from: data)
    }

    // File upload
    func uploadImage( : )   ->  {
         request  (url: baseURL.appendingPathComponent())
        request.httpMethod  

         boundary  ().uuidString
        request.setValue(, forHTTPHeaderField: )

         body  ()
        body.append()
        body.append()
        body.append()
        body.append(imageData)
        body.append()

         (data, )    session.upload(for: request, from: body)
          decoder.decode(., from: data)
    }
}

// Alamofire — request chaining
AF.request("https://api.example.com/v1/products",
           headers: ["Authorization": "Bearer \(token)"])
    .validate()
    .responseDecodable(of: [Product].self) { response in
        switch response.result {
        case .success(let products): handleProducts(products)
        case .failure(let error): handleError(error)
        }
    }

// Alamofire interceptor
class AuthInterceptor: RequestInterceptor {
    func adapt(_ urlRequest: URLRequest, for session: Session,
               completion: @escaping (Result<URLRequest, Error>) -> Void) {
        var request = urlRequest
        request.setValue("Bearer \(getToken())", forHTTPHeaderField: "Authorization")
        completion(.success(request))
    }

    func retry(_ request: Request, for session: Session, dueTo error: Error,
               completion:  () -> ) {
          statusCode  request.response.statusCode, statusCode    {
            completion(.doNotRetry)
            
        }
        refreshToken { success 
            completion(success  .retry : .doNotRetry)
        }
    }
}

iOS (Objective-C)

// NSURLSession — legacy approach
NSURLSessionConfiguration *config = [NSURLSessionConfiguration defaultSessionConfiguration];
config.timeoutIntervalForRequest = 10;
config.HTTPAdditionalHeaders = @{@"Authorization": [NSString stringWithFormat:@"Bearer %@", token]};

NSURLSession *session = [NSURLSession sessionWithConfiguration:config];

NSURL *url = [NSURL URLWithString:@"https://api.example.com/v1/products"];
NSURLSessionDataTask *task = [session dataTaskWithURL:url
    completionHandler:^(NSData *data, NSURLResponse *response, NSError *error) {
        if (error) { /* handle error */ return; }
        NSArray *products = [NSJSONSerialization JSONObjectWithData:data options:0 error:nil];
        dispatch_async(dispatch_get_main_queue(), ^{
            [self updateUIWithProducts:products];
        });
    }];
[task resume];

// AFNetworking (legacy Alamofire equivalent)
AFHTTPSessionManager *manager = [[AFHTTPSessionManager alloc] initWithBaseURL:baseURL];
[manager GET:@"products" parameters:nil headers:@{@"Authorization": authHeader}
    progress:nil
    success:^(NSURLSessionDataTask *task, id responseObject) {
        // Handle success
    }
    failure:^(NSURLSessionDataTask *task, NSError *error) {
        // Handle failure
    }];

Step 3: Implement Token Management

Auth token refresh flow (all platforms):

Request with token → 401 Unauthorized
  → Lock concurrent requests (prevent multiple refresh attempts)
  → Call refresh token endpoint
  → Success? → Update stored tokens → Retry original request + queued requests
  → Failure? → Clear tokens → Navigate to login

Flutter — Dio Token Refresh Interceptor

class AuthInterceptor extends QueuedInterceptor {
  final TokenStorage _tokenStorage;
  final Dio _refreshDio; // separate Dio instance to avoid recursion

  @override
  void onRequest(RequestOptions options, RequestInterceptorHandler handler) async {
    final token = await _tokenStorage.getAccessToken();
    if (token != null) {
      options.headers['Authorization'] = 'Bearer $token';
    }
    handler.next(options);
  }

  @override
  void onError(DioException err, ErrorInterceptorHandler handler) async {
    if (err.response?.statusCode != 401) {
      return handler.next(err);
    }

    try {
      final refreshToken = await _tokenStorage.getRefreshToken();
      final response = await _refreshDio.post('/auth/refresh', data: {
        'refresh_token': refreshToken,
      });
      await _tokenStorage.saveTokens(
        accessToken: response.data['access_token'],
        refreshToken: response.data['refresh_token'],
      );
      // Retry the failed request
      final retryResponse = await _refreshDio.fetch(err.requestOptions);
      handler.resolve(retryResponse);
    } catch (e) {
      await _tokenStorage.clearTokens();
      handler.next(err); // propagate to force logout
    }
  }
}

Android (Kotlin) — OkHttp Authenticator

class TokenAuthenticator(
    private val tokenStorage: TokenStorage,
    private val authApi: AuthApi
) : Authenticator {

    private val lock = Mutex()

    override fun authenticate(route: Route?, response: Response): Request? {
        // Max retry
        if (response.retryCount > 1) return null

        return runBlocking {
            lock.withLock {
                // Check if another thread already refreshed
                val currentToken = tokenStorage.getAccessToken()
                val requestToken = response.request.header("Authorization")?.removePrefix("Bearer ")

                if (currentToken != requestToken && currentToken != null) {
                    // Token was already refreshed by another thread
                    return@runBlocking response.request.newBuilder()
                        .header("Authorization", "Bearer $currentToken")
                        .build()
                }

                // Actually refresh
                try {
                    val refreshResponse = authApi.refresh(tokenStorage.getRefreshToken()!!)
                    tokenStorage.saveTokens(refreshResponse.accessToken, refreshResponse.refreshToken)
                    response.request.newBuilder()
                        .header("Authorization", "Bearer ${refreshResponse.accessToken}")
                        .build()
                }  (e: Exception) {
                    tokenStorage.clearTokens()
                     
                }
            }
        }
    }
}

  Response.retryCount: 
    () = generateSequence() { it.priorResponse }.count() - 

iOS (Swift) — URLSession Token Refresh

class AuthenticatedAPIClient {
    private let session: URLSession
    private let tokenStorage: TokenStorage
    private var isRefreshing = false
    private var pendingRequests: [(URLRequest, CheckedContinuation<(Data, URLResponse), Error>)] = []

    func authenticatedRequest(_ request: URLRequest) async throws -> (Data, URLResponse) {
        var authedRequest = request
        if let token = tokenStorage.accessToken {
            authedRequest.setValue("Bearer \(token)", forHTTPHeaderField: "Authorization")
        }

        let (data, response) = try await session.data(for: authedRequest)

        guard let httpResponse = response as? HTTPURLResponse else {
            return (data, response)
        }

        if httpResponse.statusCode == 401 {
             newToken    refreshTokenIfNeeded()
             retryRequest  request
            retryRequest.setValue(, forHTTPHeaderField: )
               session.data(for: retryRequest)
        }

         (data, response)
    }

      ()   ->  {
        
        
    }
}

Step 4: Implement Response Parsing

Serialization Libraries by Platform

Platform	Recommended	Alternative	Legacy
Flutter	json_serializable + freezed	built_value, manual fromJson	—
Android (Kotlin)	kotlinx.serialization	Moshi	Gson
Android (Java)	Gson	Moshi, Jackson	—
iOS (Swift)	Codable (built-in)	—	NSJSONSerialization
iOS (ObjC)	NSJSONSerialization + manual mapping	Mantle	—

Flutter — freezed + json_serializable:

@freezed
class Product with _$Product {
  const factory Product({
    required String id,
    required String name,
    required double price,
    @JsonKey(name: 'created_at') required DateTime createdAt,
    String? imageUrl,
  }) = _Product;

  factory Product.fromJson(Map<String, dynamic> json) =>
      _$ProductFromJson(json);
}

Android (Kotlin) — kotlinx.serialization:

@Serializable
data class Product(
    val id: String,
    val name: String,
    val price: Double,
    @SerialName("created_at") val createdAt: String,
    @SerialName("image_url") val imageUrl: String? = null,
)

iOS (Swift) — Codable:

struct Product: Codable {
    let id: String
    let name: String
    let price: Double
    let imageUrl: String?

    enum CodingKeys: String, CodingKey {
        case id, name, price
        case imageUrl = "image_url"
    }
}

// Custom date decoding
let decoder = JSONDecoder()
decoder.dateDecodingStrategy = .iso8601
decoder.keyDecodingStrategy = .convertFromSnakeCase

Step 5: Implement Caching & Offline Support

Strategy	Implementation	When to Use
HTTP caching (ETag/Cache-Control)	Handled by HTTP client + server headers	Read-heavy, data changes infrequently
In-memory cache	LRU cache in repository layer	Session-scoped, frequently accessed data
Database cache	Room / Core Data / Drift with TTL	Offline-capable, structured data
Offline queue	Queue writes, sync when online	Offline-first apps, unreliable networks

Repository pattern with cache:

API Request
  → Check in-memory cache (instant)
  → Check database cache (fast, check TTL)
  → Fetch from network (slow)
  → Update database cache
  → Update in-memory cache
  → Return data

Flutter — Cache-first with Dio:

class ProductRepository {
  final ProductApi _api;
  final ProductDao _dao;
  final Duration _cacheTTL = const Duration(minutes: 15);

  Future<List<Product>> getProducts({bool forceRefresh = false}) async {
    if (!forceRefresh) {
      final cached = await _dao.getAll();
      if (cached.isNotEmpty && !_isStale(cached.first.cachedAt)) {
        return cached;
      }
    }

    try {
      final products = await _api.getProducts();
      await _dao.replaceAll(products);
      return products;
    } catch (e) {
      // Fallback to cache on network error
      final cached = await _dao.getAll();
      if (cached.isNotEmpty) return cached;
      rethrow;
    }
  }
}

Step 6: Handle Errors Consistently

Error model (all platforms):

NetworkError
  ├── NoConnection          → show offline banner
  ├── Timeout               → retry with backoff
  ├── ServerError (5xx)     → retry with backoff, show generic error
  ├── ClientError (4xx)
  │   ├── Unauthorized (401) → refresh token or logout
  │   ├── Forbidden (403)    → show permission error
  │   ├── NotFound (404)     → show not found
  │   └── Validation (422)   → show field-level errors
  └── ParseError            → log, show generic error

Flutter:

sealed class NetworkError {
  const NetworkError();
}
class NoConnection extends NetworkError {}
class Timeout extends NetworkError {}
class ServerError extends NetworkError { final int code; const ServerError(this.code); }
class Unauthorized extends NetworkError {}
class ValidationError extends NetworkError { final Map<String, List<String>> fields; const ValidationError(this.fields); }
class Unknown extends NetworkError { final String message; const Unknown(this.message); }

NetworkError mapDioError(DioException e) => switch (e.type) {
  DioExceptionType.connectionTimeout => const Timeout(),
  DioExceptionType.connectionError => const NoConnection(),
  _ => switch (e.response?.statusCode) {
    401 => const Unauthorized(),
    422 => ValidationError(e.response?.data['errors']),
    >= 500 => ServerError(e.response!.statusCode!),
    _ => Unknown(e.message ?? 'Unknown error'),
  },
};

Step 7: WebSocket / Real-Time Connections

Platform	Library	Use Case
Flutter	web_socket_channel, socket_io_client	Chat, live updates, notifications
Android	OkHttp WebSocket, Scarlet, Ktor WebSocket	Chat, live updates, notifications
iOS	URLSessionWebSocketTask, Starscream	Chat, live updates, notifications

Flutter — WebSocket:

final channel = WebSocketChannel.connect(Uri.parse('wss://api.example.com/ws'));

channel.stream.listen(
  (message) { /* handle incoming message */ },
  onError: (error) { /* reconnect with backoff */ },
  onDone: () { /* connection closed, reconnect */ },
);

channel.sink.add(jsonEncode({'type': 'subscribe', 'topic': 'orders'}));

Output Format

## Networking Architecture
- **Platform:** [Flutter / Android / iOS]
- **HTTP Client:** [Dio / Retrofit+OkHttp / URLSession / Alamofire]
- **Auth:** [Bearer JWT / OAuth2 / API key]
- **Serialization:** [freezed / kotlinx.serialization / Codable]
- **Caching:** [HTTP cache / Database / In-memory / Offline queue]

## API Layer Structure
[Class diagram or module structure]

## Token Management
[Auth flow diagram]

## Error Handling
[Error type mapping and UI behavior]

## Caching Strategy
[What is cached, TTL, invalidation]

Quality Checklist

• HTTP client is configured with timeouts (connect, read, write)
• Auth token refresh handles concurrent requests (no race conditions)
• Certificate pinning is configured (see /mobile-security)
• Error responses are mapped to domain error types (not raw HTTP errors in UI)
• Network calls are NOT on the main thread
• Logging is enabled in debug, disabled in release
• Retry logic uses exponential backoff with jitter
• Request cancellation is handled (screen dismissed, search debounced)
• File uploads show progress
• Offline fallback returns cached data when network fails

Edge Cases

• Token refresh race condition: multiple concurrent 401s should coalesce into a single refresh, not multiple
• Large response bodies should be streamed, not loaded entirely into memory
• Untrusted server certificates in development: use a debug-only network config, never disable SSL in production
• GraphQL error responses return 200 with errors in the body — check data.errors even on 200
• WebSocket reconnection needs exponential backoff to avoid thundering herd on server recovery
• On Android, Dispatchers.IO is limited to 64 threads — for many concurrent requests, use a custom dispatcher
• On iOS, URLSession background sessions continue uploads/downloads when the app is suspended