The starting address: www.ljcoder.com/16080130499…
RxSwift operator
Operators can help you create new sequences, or combine existing sequences to generate a new sequence. For example, filter and map conversion, see the documentation operator-operator.
Operators such as filter and map are higher-order functions that take one or more functions as input and output a function. Let’s take MAP as an example for analysis.
Analysis of MAP Principle
Let’s start with the simplest map usage code
let ob = Observable.of(1.2.3.4)
ob.map { (number) -> Int in
return number+2
}
.subscribe{
print("\ [$0)")
}
.disposed(by: disposeBag)
Copy the code
To viewmap
Function, as you can see in the comments section, is implemented in the map file
// Map.swift
extension ObservableType {
public func map<Result> (_ transform: @escaping (Element) throws -> Result)
-> Observable<Result> {
return Map(source: self.asObservable(), transform: transform)
}
}
Copy the code
In RxSwift core logic analysis, AnonymousObservable is also a subclass of Producer. There should be some similarities between them. Let’s analyze the differences.
final private class Map<SourceType.ResultType> :Producer<ResultType> {
typealias Transform = (SourceType) throws -> ResultType
private let source: Observable<SourceType>
private let transform: Transform
init(source: Observable<SourceType>, transform: @escaping Transform) {
self.source = source
self.transform = transform
}
override func run<Observer: ObserverType> (_ observer: Observer.cancel: Cancelable) -> (sink: Disposable, subscription: Disposable) where Observer.Element = = ResultType {
let sink = MapSink(transform: self.transform, observer: observer, cancel: cancel)
let subscription = self.source.subscribe(sink)
return (sink: sink, subscription: subscription)
}
}
Copy the code
- Much more
source
Property to store the original queue,transform
Store the transformation logic, which is.map
Closure in the backreturn number+2
. sink
Also byAnonymousObservableSink
Turned out to beMapSink
.subscription
issource
thesubscribe
The return value.
As we already know from RxSwift- core logic analysis, MapSink’s on(_:) method is finally called.
final private class MapSink<SourceType.Observer: ObserverType> :Sink<Observer>, ObserverType {
typealias Transform = (SourceType) throws -> ResultType
typealias ResultType = Observer.Element
typealias Element = SourceType
private let transform: Transform
init(transform: @escaping Transform.observer: Observer.cancel: Cancelable) {
self.transform = transform
super.init(observer: observer, cancel: cancel)
}
func on(_ event: Event<SourceType>) {
switch event {
case .next(let element):
do {
let mappedElement = try self.transform(element)
self.forwardOn(.next(mappedElement))
}
catch let e {
self.forwardOn(.error(e))
self.dispose()
}
case .error(let error):
self.forwardOn(.error(error))
self.dispose()
case .completed:
self.forwardOn(.completed)
self.dispose()
}
}
}
Copy the code
Next, element is the element of the source sequence (1,2,3,4), transform is called, number+2, and the transformed element is sent, and the map is complete.
// If the sequence element is 1, the sequence element is 3 after number+2
let mappedElement = try self.transform(element)
// If you send 3, the subscribed element becomes 3
self.forwardOn(.next(mappedElement))
Copy the code
conclusion
- The subscription and response logic of higher-order functions is the same as that of ordinary sequences.
- Different higher-order functions have different logic
Producer
Subclasses and differentSink
Subclasses are processed to ensure the flexibility and extensibility of the sequence.