Welcome to ReCpp, a sophisticated C++ library for reactive programming that builds on the Reactive Streams standard and leverages the ReactiveX model.
This library, powered by the RsCpp implementation of the Reactive Streams standard, empowers developers to create clean, concise, and efficient asynchronous and event-driven code with ease.
-
Reactive Streams Standard: ReCpp is built on top of the RsCpp implementation of the Reactive Streams standard, providing a consistent and interoperable API for handling asynchronous data streams.
-
Functional Reactive Programming (FRP): Embrace the paradigm of FRP to model complex asynchronous workflows with a clear and functional approach.
-
Comprehensive Operators: Enjoy a rich set of operators inspired by the ReactiveX model, enabling effortless manipulation and transformation of data streams.
-
Backpressure Handling: Efficiently manage backpressure to ensure your application remains responsive and scalable, even in high-load scenarios.
-
Extensibility: Easily extend the library with your own custom operators or integrate it with existing codebases.
ReCpp avoid the callback hell, when multiple callbacks are nested into each other. ReCpp allows you to chain your callbacks in a linear way and handle errors more easily. For example :
// Callback hell
asyncFunction1(
[]()
{
asyncFunction2(
[]()
{
doSomething();
},
[](const auto &error)
{
std::cerr << error.what() << std::endl;
});
},
[](const auto &error)
{
std::cerr << error.what() << std::endl;
});
// ReCpp chaining
rxAsyncFunction1()
.andThen(rxAsyncFunction2())
.subscribe(
[]()
{
doSomething();
},
[](const auto &error)
{
std::cerr << error.what() << std::endl;
});Even if c++20 coroutines offers a really great standard solution for asynchronous programming, ReCpp has some advantages over them:
- natively forward errors which leads to a higher resiliency
- faster to write
- easier to read
- bring powerful operators to create highly performant asynchronous code that would be hard to create with coroutines
- backpressure handling
- Windows
- MacOS
- Ubuntu
You want to integrate ReCpp to your project ? Check our installation guide
Explore the Documentation to learn more about ReCpp.
ReCpp can easily interoperate with other asynchronous programming systems. Here are some examples:
// The async function that uses callbacks
void asyncFunction(const std::function<void(int)> &onSuccess, const std::function<void(const std::exception_ptr &)> onError);
// The ReCpp equivalent
recpp::Single<int> rxAsyncFunctionWrapper()
{
return recpp::Single<int>::create(
[](auto &subscriber) // Lambda called on subscription
{
asyncFunction(
[subscriber](int result) // Lambda called on asyncFunction success
{
subscriber.onNext(result); // Emit the value and complete the Single
},
[](const auto &error) // Lambda called on asyncFunction error
{
subscriber.onError(error); // Emit the error
});
});
}// The async function that uses coroutines
task<std::expected<int, std::exception_ptr>> asyncFunction();
// The ReCpp equivalent
recpp::Single<int> rxAsyncFunctionWrapper()
{
return recpp::Single<int>::create(
[](auto &subscriber) // Lambda called on subscription
{
const auto result = asyncFunction().get();
if (result.has_value())
subscriber.onNext(result.value());
else
subscriber.onError(result.error());
});
}// The rxcpp function
rxcpp::observable<int> rxcppAsyncFunction();
// The ReCpp equivalent
recpp::Observable<int> recppAsyncFunction()
{
return recpp::Observable<int>::create(
[](auto &subscriber) // Lambda called on subscription
{
rxcppAsyncFunction()
.subscribe(
[subscriber](auto value)
{
subscriber.onNext(value);
},
[subscriber](auto error)
{
subscriber.onError(error);
},
[subscriber]()
{
subscriber.onComplete();
}
);
});
}ReCpp is licensed under the MIT License. See the LICENSE file for details.
We welcome contributions from the community! Whether you want to report a bug, request a feature, or submit a pull request, please follow our Contribution Guidelines.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent