vexana 3.4.2

Manage your network layer with dio. You will be able to use it very easily network operations.

vexana

Before Starting MY COUNTRY NEEDS TO HELP BECAUSE EARTHQUAKE 7.8 and 7.6 within some day!

Turkey has recently been struck by a devastating earthquake with a magnitude of 7.8. The impact has been widespread and many communities have been severely affected. In this difficult time, we are reaching out to ask for your support. Any donation, no matter how small, will help us provide much-needed aid to those affected by the earthquake. Your generosity will help provide shelter, food, and medical supplies to those in need. Your contribution will make a real difference in the lives of those affected by this disaster. Thank you for your kindness and support.

You can donate for AHBAP with this site AHBAP_DONATE

Vexana is easy to use network process with dio. You can do dynamic model parsing, base error model, timeout and many utility functions.

Getting Started 🔥

Let's talk about usage details.

Network Manager 😎

Have a lot of options: baseurl, logger, interceptors, base model etc.

If you want to manage your error model, you just declare your model so it's way getting the error model everywhere.

INetworkManager  networkManager = NetworkManage<Null or UserErrorModel>(isEnableLogger: true, errorModel: UserErrorModel(),
 options: BaseOptions(baseUrl: "https://jsonplaceholder.typicode.com/"));

Model Parse ⚔️

First, you have to provide the parse model, then the result model. (Result model could be a list, model or primitive)

final response =
await networkManager.send<Todo, List<Todo>>("/todos", parseModel: Todo(), method: RequestType.GET);

Base Headers 📍

You could be add key values to your every request directly.(add authentication key)

networkManager.addBaseHeader(MapEntry(HttpHeaders.authorizationHeader, response.data?.first.title ?? ''));
// Clear single header
networkManager.removeHeader('\${response.data?.first.id}');
// Clear all header
networkManager.clearHeader();

Download File 📍

Download File Simple

You can download any file format like pdf, png or etc with progress indicator.

final response = await networkManager.downloadFileSimple('http://www.africau.edu/images/default/sample.pdf', (count, total) {
      print('${count}');
});

//Example: Image.memory(response.data)

Download File

You can download by specifying model and request type.

final response = await networkManager.downloadFile(
    'financial-report',
    (count, total) {
      print('${count}');
    },
    method: RequestType.POST,
    data: FileDownloadModel(),
);

HTTP Post Request with Request Body 🚀

The model found in the request body must extend the abstract class INetworkModel, as follows.

class TodoPostRequestData extends INetworkModel<TodoPostRequestData>

Then, since the model will have toJson and fromJson properties, you can create the object and pass it directly to the send method.

So, it is sufficient to send only the request body object into the send method. You don't need to use toJson.

final todoPostRequestBody = TodoPostRequestData();
final response =
await networkManager.send<Todo, List<Todo>>("/todosPost", parseModel: Todo(), method: RequestType.POST, data: todoPostRequestBody);

Cancel Request ❌

You can implement cancel token when need to invoke your request during to complete.

  final cancelToken = CancelToken();
    networkManager
        .send<ReqResModel, ReqResModel>('/users?delay=5',
            parseModel: ReqResModel(), method: RequestType.GET, canceltoken: cancelToken)
        .catchError((err) {
      if (CancelToken.isCancel(err)) {
        print('Request canceled! ' + err.message);
      }
    });

    cancelToken.cancel('canceled');

    await Future.delayed(const Duration(seconds: 8));

Primitive Request 🌼

Sometimes we need to parse only primitive types for instance List

//
[
  "en",
  "tr",
  "fr"
]
//
networkManager.sendPrimitive<List>("languages");

Network Model 🛒

You must wrap your model with INetworkModel so that, we understand model has toJson and fromJson methods.

class Todo extends INetworkModel<Todo>

Refresh Token ♻️

Many projects use authentication structure for mobile security (like a jwt). It could need to renew an older token when the token expires. This time provided a refresh token option, we can lock all requests until the token refresh process is complete.

Since i locked all requests, I am giving a new service instance.

INetworkManager  networkManager = NetworkManager(isEnableLogger: true, options: BaseOptions(baseUrl: "https://jsonplaceholder.typicode.com/"),
onRefreshFail: () {  //Navigate to login },
 onRefreshToken: (error, newService) async {
    <!-- Write your refresh token business -->
    <!-- Then update error.req.headers to new token -->
    return error;
});

Caching 🧲

You need to write a response model in the mobile device cache sometimes. It's here now. You can say expiration date and lay back 🙏

    await networkManager.send<Todo, List<Todo>>("/todos",
        parseModel: Todo(),
        expiration: Duration(seconds: 3),
        method: RequestType.GET);

You must declare a caching type. It has FileCache and SharedCache options now. NetworkManager(fileManager: LocalFile()); If you want more implementation details about the cache, you should read this article

Without Network connection 🧲

Especially, mobile device many times lost connection for many reasons so if you want to retry your request, you need to add this code and that's it. Your app user can be show bottom sheet dialog and they will be use this features only tree times because i added this rule.

    // First you must be initialize your context with NoNetwork class
    networkManager = NetworkManager(
      isEnableLogger: true,
      noNetwork: NoNetwork(context),
      options: BaseOptions(baseUrl: 'https://jsonplaceholder.typicode.com'),

      errorModelFromData: _errorModelFromData, //This is optional.
    );

    // If you want to create custom widget, you can add in no network class with callback function.
      networkManager = NetworkManager(
      isEnableLogger: true,
      noNetwork: NoNetwork(
        context,
        customNoNetwork: (onRetry) {
          // You have to call this retry method your custom widget
          return NoNetworkSample(onPressed: onRetry);
        },
      ),
      options: BaseOptions(baseUrl: 'https://jsonplaceholder.typicode.com'),

      //Example request
       final response = await networkManager.send<Post, List<Post>>('/posts',
        parseModel: Post(), method: RequestType.GET, isErrorDialog: true);

And result!!

Error model handle ❎

This point so important for many apps. Some business operation want to show any message or logic when user did a mistake like wrong password etc. You can manage very easily to error model for whole project with this usage.

INetworkManager  networkManager = NetworkManage<UserErrorModel>(isEnableLogger: true, errorModel: UserErrorModel(),
 options: BaseOptions(baseUrl: "https://jsonplaceholder.typicode.com/"));

 IResponseModel<List<Post>?, BaseErrorModel?> response =  networkManager.send<Post, List<Post>>('/posts',
        parseModel: Post(), method: RequestType.GET);
      <!-- Error.model came from your backend with your declaration -->
      showDialog(response.error?.model?.message)

    response

Tasks

---

• ✅ Example project
• ✅ Unit Test with json place holder
• ✅ Unit Test with custom api
• ✅ Handle network problem
• ❌ Make a unit test all layers.
• ✅ Cache Option
  • ❌ ISAAR Support
  • ✅ Web Cache Support
• ✅ Refresh Token Architecture
• ❌ Usage Utility
• ❌ Readme Update

License

2020 created for @VB10

Youtube Channel

---

Contributors

Made with contrib.rocks.