I have a 3rd-party SDK where architecture is like this:
I call a method of the service (lets call it TheirService)
Once that method completes, a delegate of that service (lets call it TheirServiceDelegate will receive a notification onSuccess if call succeeds or onFailure if it fails.
The problem with that is that methods of TheirService are dependant on each other, so I end up having a "chain reaction", where each next method of TheirService is called from previous callback.
Note that TheirService is single threaded, and is really picky about previous method to be complete before I can start the next one.
Currently my interactions with the service look like this:
protocol MyClientListener {
notifyOnFailure()
notifyOnResult(result: SomeObject)
}
class MyClient: TheirServiceDelegate {
static let instance = MyClient()
let myService = TheirService()
var myResult: SomeObject?
var resultListener: MyClientListener
private init() { }
func start(resultListener: MyClientListener) {
self.resultListener = resultListener
myService.initialize()
}
// TheirServiceDelegate method
func onInitializeSuccess() {
myService.bootstrap()
}
// TheirServiceDelegate method
func onBootstrapSuccess() {
myService.login(user, password)
}
// TheirServiceDelegate method
func onLoginSuccess() {
myService.doSomethingUsefulStep1()
}
// TheirServiceDelegate method
func onDoSomethingUsefulStep1Success() {
myService.doSomethingUsefulStep2()
}
// TheirServiceDelegate method
func onDoSomethingUsefulStep2Success(result: SomeObject) {
// ah, look, now I have some object I actually wanted!
resultListener.notifyOnResult(result)
}
}
I also have to deal with failure case for each of them. And I cannot skip or change the order of steps, which creates some sort of awkward state machine.
Instead i would like to interact with service via logical functions, that complete certain stages of the process from end to end, waiting for results between the steps:
class MyClient {
static let instance = MyClient()
let myService = MyService()
private init() { }
func connect() throws {
myService.initialize()
// wait for success or failure, throw on failure
myService.bootstrap()
// wait for success or failure, throw on failure
myService.login(user, password)
// wait for success or failure, throw on failure
}
func doSomethingUseful() -> SomeObject {
myService.doSomethingUsefulStep1()
// wait for success or failure, throw on failure
myService.doSomethingUsefulStep2()
// wait for success or failure, throw on failure
// on success, it will get an object it could return
}
Called like this:
try MyClient.instance.connect()
let x = try MyClient.instance.doSomethingUseful()
So is there any way to turn "wait for success or failure" comment into actual code that waits for that single-threaded service to call back? And where a delegate would fit in that case?
Related
I need to support cancellation of a function that returns an object that can be cancelled after initiation. In my case, the requester class is in a 3rd party library that I can't modify.
actor MyActor {
...
func doSomething() async throws -> ResultData {
var requestHandle: Handle?
return try await withTaskCancellationHandler {
requestHandle?.cancel() // COMPILE ERROR: "Reference to captured var 'requestHandle' in concurrently-executing code"
} operation: {
return try await withCheckedThrowingContinuation{ continuation in
requestHandle = requester.start() { result, error in
if let error = error
continuation.resume(throwing: error)
} else {
let myResultData = ResultData(result)
continuation.resume(returning: myResultData)
}
}
}
}
}
...
}
I have reviewed other SO questions and this thread: https://forums.swift.org/t/how-to-use-withtaskcancellationhandler-properly/54341/4
There are cases that are very similar, but not quite the same. This code won't compile because of this error:
"Reference to captured var 'requestHandle' in concurrently-executing code"
I assume the compiler is trying to protect me from using the requestHandle before it's initialized. But I'm not sure how else to work around this problem. The other examples shown in the Swift Forum discussion thread all seem to have a pattern where the requester object can be initialized before calling its start function.
I also tried to save the requestHandle as a class variable, but I got a different compile error at the same location:
Actor-isolated property 'profileHandle' can not be referenced from a
Sendable closure
You said:
I assume the compiler is trying to protect me from using the requestHandle before it’s initialized.
Or, more accurately, it is simply protecting you against a race. You need to synchronize your interaction with your “requester” and that Handle.
But I’m not sure how else to work around this problem. The other examples shown in the Swift Forum discussion thread all seem to have a pattern where the requester object can be initialized before calling its start function.
Yes, that is precisely what you should do. Unfortunately, you haven’t shared where your requester is being initialized or how it was implemented, so it is hard for us to comment on your particular situation.
But the fundamental issue is that you need to synchronize your start and cancel. So if your requester doesn’t already do that, you should wrap it in an object that provides that thread-safe interaction. The standard way to do that in Swift concurrency is with an actor.
For example, let us imagine that you are wrapping a network request. To synchronize your access with this, you can create an actor:
actor ResponseDataRequest {
private var handle: Handle?
func start(completion: #Sendable #escaping (Data?, Error?) -> Void) {
// start it and save handle for cancelation, e.g.,
handle = requestor.start(...)
}
func cancel() {
handle?.cancel()
}
}
That wraps the starting and canceling of a network request in an actor. Then you can do things like:
func doSomething() async throws -> ResultData {
let responseDataRequest = ResponseDataRequest()
return try await withTaskCancellationHandler {
Task { await responseDataRequest.cancel() }
} operation: {
return try await withCheckedThrowingContinuation { continuation in
Task {
await responseDataRequest.start { result, error in
if let error = error {
continuation.resume(throwing: error)
} else {
let resultData = ResultData(result)
continuation.resume(returning: resultData)
}
}
}
}
}
}
You obviously can shift to unsafe continuations when you have verified that everything is working with your checked continuations.
After reviewing the Swift discussion thread again, I see you can do this:
...
var requestHandle: Handle?
let onCancel = { profileHandle?.cancel() }
return try await withTaskCancellationHandler {
onCancel()
}
...
I tried to put this example into a simple project + ViewController but can't get it to compile
https://www.hackingwithswift.com/quick-start/concurrency/how-to-get-a-result-from-a-task
I'm trying to call fetchQuotes() from an IBAction via a button tap, however because fetchQuotes is marked async, I'm getting an error.
I know that I can wrap the call to fetchQuotes() in a task:
Task {
fetchQuotes()
}
, but this doesn't make sense to me since fetchQuotes is already creating tasks.
Can anyone advise?
Here is my code:
// https://www.hackingwithswift.com/quick-start/concurrency/how-to-get-a-result-from-a-task
import UIKit
class ViewController: UIViewController {
override func viewDidLoad() {
super.viewDidLoad()
// Do any additional setup after loading the view.
}
#IBAction func buttonTap(_ sender: Any) {
fetchQuotes()
}
func fetchQuotes() async {
let downloadTask = Task { () -> String in
let url = URL(string: "https://hws.dev/quotes.txt")!
let data: Data
do {
(data, _) = try await URLSession.shared.data(from: url)
} catch {
throw LoadError.fetchFailed
}
if let string = String(data: data, encoding: .utf8) {
return string
} else {
throw LoadError.decodeFailed
}
}
let result = await downloadTask.result
do {
let string = try result.get()
print(string)
} catch LoadError.fetchFailed {
print("Unable to fetch the quotes.")
} catch LoadError.decodeFailed {
print("Unable to convert quotes to text.")
} catch {
print("Unknown error.")
}
}
}
enum LoadError: Error {
case fetchFailed, decodeFailed
}
Here's a screenshot of my code + Xcode error:
this doesn't make sense to me since fetchQuotes is already creating tasks
What fetchQuotes is internally doing is an implementation detail, it might as well not explicitly create any tasks, or not even be doing async calls.
Now, the async part of the method declaration tells the compiler the method might suspend the caller thread, thus any caller needs to provide an async context when calling the method. Note that I said might suspend not will suspend, this is a subtle, but important difference.
The tasks created by the method are valid only during the method execution, so if you need to be able to call the method from a non-async context you'll have to create new tasks.
However, you can tidy-up your code by providing a non-async overload:
func fetchQuotes() {
Task {
await fetchQuotes()
}
}
, which can be called from your #IBAction
#IBAction func buttonTap(_ sender: Any) {
fetchQuotes()
}
There's no escaping out of this, if you want to be able to call an async function from a non-async context, then you'll need a Task, regardless on how the called method is implemented. Structured concurrency tasks are not reusable.
but this doesn't make sense to me since fetchQuotes is already creating tasks
Irrelevant; you've internalized a false rule, or failed to internalize the real rule. This has nothing to do with who is "creating tasks". It has to do with the real rule, the most fundamental rule of Swift's async/await, which is:
You can only call an async method from within an async context.
Well, the method buttonTap is not an async method, so your call to fetchQuotes is not being made within an async context. That's illegal, and the compiler stops you dead.
But if you wrap the call to fetchQuotes in a Task initializer, that is an async context. Success!
Basically, the effect you're seeing is what I call the regress of doom. Since every call to an async method must be in an async context, but since your events from Cocoa, such as an IBAction, are not async, how can your code ever call an async method? The answer is a Task initializer: it gives you a place to get started.
I have a singleton class, which requires a network operation with asynchronous callback to initialize. Any member function of that class should only be called when initialization callback finished.
What I'm not clear with, is how to wait for asynch callback to finish before I return a singleton instance to a caller? Or maybe is there any other way to ensure initialization completes before any function can be called (only initialization should be sequential, all operations after that shouldn't be).
Here's the minimal code:
class DataProvider {
public static let instance: DataProvider {
// on first call wait here until callback is done
// on subsequent calls, no need to wait since already initialized
return internalInstance
}
private static let internalInstance = DataProvider()
private init() {
initialize()
}
private func initialize() {
Something.callAsynch { (result, error) in
// instance becomes ready when this line is executed
}
}
public func doSomething() {
// this function should only run after asynch callback was executed
}
}
This class is called by other classes and potentially other threads like this:
DataProvider.instance.doSomething()
With some help, I found a good solution, that is both reliable and simple. The solution is to use count-down latch. Unfortunately there's no built-in class like that in Swift, but some implementations could be found online (for example: this class from Uber). So this is a solution:
class DataProvider {
public static let instance: DataProvider {
// on first call wait here until callback is done
// on subsequent calls, no need to wait since already initialized
return internalInstance
}
private static let internalInstance = DataProvider()
private let initLatch = CountDownLatch(1)
private init() {
initialize()
}
private func initialize() {
Something.callAsynch { (result, error) in
// instance becomes ready when this line is executed
initLatch.countDown()
}
}
public func doSomething() {
initLatch.await()
// this function should only run after asynch callback was executed
}
}
So what does it do:
the latch is set to 1 so any caller of doSomething will be suspended until latch is decremented to 0
latch is only decremented to 0 in one place (init completion callback), hence nobody will proceed before initialization completed
but after latch is set to 0, calling doSomething will not be delayed
Notes:
This is a simplified model for problem and solution. In reality there are some other factors that may need to be taken into account, and would require other concurrency tools.
In my case, the callers are non-interactive, hence I do want them to wait rather than be asynchronous. If callers were interactive, I'd let them deal with asynchronous nature of this class instead.
I'm trying to create an online mobile application and can't figure out the best way to handle functions with multiple asynchronous calls. Say I have a function for example that updates a user in some way, but involved multiple asynchronous calls in the single function call. So for example:
// Function caller
update(myUser) { (updatedUser, error) in
if let error = error {
// Present some error UI to the user
}
if let updatedUser = updatedUser {
// Do something with the user
}
}
// Function implementation
public func updateUser(user: User, completion: #escaping (User?, Error?) -> () {
// asynchronous call A
updateUserTable(user: User) { error in
if let error = error {
completion(nil, error)
} else {
// create some new user object
completion(user, nil)
}
}
// asynchronous call B
uploadMediaForUser(user: User) { error in
if let error = error {
completion(nil, error)
}
}
// asynchronous call C
removeOldReferenceForUser(user: User) { error in
if let error = error {
completion(nil, error)
}
}
// Possibly any additional amount of asynchronous calls...
}
In a case like this, where one function call like updating a user involved multiple asynchronous calls, is this an all or nothing situation? Say for example the updateUserTable() call completes, but the user disconnects from the internet as uploadMediaForUser() was running, and that throws an error. Since updateUserTable() completed fine, my function caller thinks this method succeeded when in fact not everything involved in updating the user completed. Now I'm stuck with a user that might have mismatched references or wrong information in my database because the user's connection dropped mid call.
How do I handle this all or nothing case? If EVERY asynchronous call completed without an error, I know updating the user was a success. If only a partial amount of asynchronous calls succeeded and some failed, this is BAD and I need to either undo the changes that succeeded or attempt the failed methods again.
What do I do in this scenario? And also, and how do I use my completion closure to help identify the actions needed depending on the success or failure of the method. Did all them succeed? Good, tell the user. Do some succeed and some failed? Bad, revert changes or try again (i dont know)??
Edit:
Just calling my completion with the error doesn't seem like enough. Sure the user sees that something failed, but that doesn't help with the application knowing the steps needed to fix the damage where partial changes were made.
I would suggest adding helper enums for your tasks and returned result, things like (User?, Error?) have a small ambiguity of the case when for example both are nil? or you have the User and the Error set, is it a success or not?
Regarding the all succeeded or some failed - I would suggest using the DispatchGroup to notify when all tasks finished (and check how they finished in the end).
Also from you current code, when some request fails it's not clear for which user - as you pass nil, so it might bring difficulties in rolling it back after failure.
So in my point of view something like below (not tested the code, but think you should catch the idea from it) could give you control about the issues you described:
public enum UpdateTask {
case userTable
case mediaUpload
// ... any more tasks you need
}
public enum UpdateResult {
case success
case error([UpdateTask: Error])
}
// Function implementation
public func updateUser(user: User, completion: #escaping (User, UpdateResult) -> ()) {
let updateGroup = DispatchGroup()
var tasksErrors = [UpdateTask: Error]()
// asynchronous call A
updateGroup.enter()
updateUserTable(user: User) { error in
if let error = error {
tasksErrors[.userTable] = error
}
updateGroup.leave()
}
// ... any other similar tasks here
updateGroup.notify(queue: DispatchQueue.global()) { // Choose the Queue that suits your needs here by yourself
if tasksErrors.isEmpty {
completion(user, .success)
} else {
completion(user, .error(tasksErrors))
}
}
}
Keep a “previous” version of everything changed, then if something failed revert back to the “previous” versions. Only change UI once all returned without failure, and if one failed, revert to “previous” version.
EX:
var temporary = “userName”
getChanges(fromUser) {
If error {
userName = temporary //This reverts back due to failure.
}
}
I am working on testing an API through Alamofire. I need to make a single call to the server to prepare it for the integration test. Once that is done, I am ready to start running tests.
The usual override setUp() is run for every test, so I do not want to do that.
I have therefore chosen to override the class setUp() as described here: https://developer.apple.com/reference/xctest/xctestcase
That's all well and good, but now, I no longer can use the standard waitForExpectations. (In the class override setUp()) I get several compiler errors that tell me that I am no longer calling the same waitForExpectations because I am in a class method, not a test case.
To try to get around this, I wanted to use a semaphore like so:
class ServiceLayerTests: XCTestCase {
static var apiService: APIService = APIService()
let sessionManager = SessionManager(serverTrustPolicyManager: ServerTrustPolicyManager(policies: ["http://localhost:3000/": .disableEvaluation]))
static var companyManger: UserWebResource?
static var companyManagerID = -1
override class func setUp() {
apiService.baseURL = "http://localhost:3000/"
beginIntegrationTests()
}
class func beginIntegrationTests() {
var semaphore = DispatchSemaphore(value: 0)
apiService.beginIntegrationTests(completion: {resource, error in
if let resource = resource {
if let manager = resource as? UserWebResource {
companyManger = manager
companyManagerID = manager.id
semaphore.signal()
}
}
})
_ = semaphore.wait(timeout: DispatchTime.distantFuture)
}
}
This does not work. Under the hood, there is an alamo fire call to the server and it responds with the user to use for the integration tests. I do see the server spinning, so I know that the actual communication is happening, but I never get into the completion closure.
I suspect I am not understanding how Swift does semaphores and that I have created a deadlock somehow. If somebody has a better solution, I'd be more than happy to hear it.
I get several compiler errors that tell me that I am no longer calling
the same waitForExpectations because I am in a class method, not a
test case
That makes sense. What you probably want is to refactor so that you are in a test case:
override class func setUp() {
apiService.baseURL = "http://localhost:3000/"
}
func testIntegrationTests() {
let urlExpectation = expectation(description: "INTEGRATION TEST")
apiService.beginIntegrationTests(completion: {resource, error in
// ...
urlExpectation.fulfill()
})
// not sure what an acceptable timeout would be, I chose this at random
waitForExpectations(timeout: 25) { error in
if let error = error {
print("Error: \(error.localizedDescription)")
}
}
}
One of the best resources with some good test examples can be found here: http://nshipster.com/xctestcase/
You can create the expectation as a lazy var that executes your one-time set up and fulfills on completion.
At the beginning of your per-test setUp() function you can wait for that expectation.
None of your tests will run until it is fulfilled, and the initial setup will run only once.
class WaitForSetup_Tests: XCTestCase {
lazy var initialSetupFinished: XCTestExpectation = {
let initialSetupFinished = expectation(description: "initial setup finished")
initialSetupTask() { // your setup task that executes this closure on completion
initialSetupFinished.fulfill()
return
}
return initialSetupFinished
}()
override func setUp() {
wait(for: [initialSignOutFinished], timeout: 2.0)
// do your per-test setup here
}
}
Note: This solution avoids using the override class function setUp() class method, because I couldn't figure out how to use the expectations except for in an instance.