Swift 3 pass data through completion with userID - ios

I'm trying to run a check on Firebase to see if a user exists, then I need to check for specific vales before continuing. I currently have this:
func myFirebaseNetworkDataRequest(finished: () -> Void) {
if let user = FIRAuth.auth()?.currentUser {
self.userUUID = (user.uid)
getUser(userUUID: self.userUUID)
finished()
}
}
In my view did load:
myFirebaseNetworkDataRequest {
// perform further operations here after data is fetched
if AppState.sharedInstance.user == true {
//present 1st view controller
} else {
//present 2nd view controller
}
In my "getUser" function:
func getUser(userUUID: String) {
let userFacebookRef = FIRDatabase.database().reference(withPath: "users").child(userUUID)
//The rest of the Firebase function.
AppState.sharedInstance.user == results.active
//active is = to true
What currently happens is that if presents the 2nd view controller because firebase hasent finished yet. I realize I need a block because firebase is already asnyc but how do I send userUUID through the closure/block?

You can have your closure parameter have an parameter of it's own. Something like,
func myFirebaseNetworkDataRequest(finished: (_ isAuthenticated: Bool) -> Void)
Then in your viewDidLoad, you would make your request...
open override func viewDidLoad() {
self.myFirebaseNetworkDataRequest( (isAuthenticated) -> {
if isAuthenticated {
// Present VC1
} else {
// Present VC 2
}
}
}

Related

In Cognito on iOS, handling new password required doesn't ever reach didCompleteNewPasswordStepWithError

I'm trying to implement functionality to respond to FORCE_CHANGE_PASSWORD on my iOS app that uses AWS Cognito. I used this Stack Overflow question which references this sample code. Right now, my code opens a view controller like it's supposed to; however, once on that view controller, I can't get it to do anything. In the sample code, it seems that when you want to submit the password change request you call .set on an instance of AWSTaskCompletionSource<AWSCognitoIdentityNewPasswordRequiredDetails>, yet when I do this, the protocol function didCompleteNewPasswordStepWithError is never called. Interestingly, the other protocol function getNewPasswordDetails is called quickly after viewDidLoad and I can't tell why. I believe this shouldn't be called until the user has entered their new password, etc and should be in response to .set but I could be wrong.
My code is pretty identical to the sample code and that SO post, so I'm not sure what's going wrong here.
My relevant AppDelegate code is here:
extension AppDelegate: AWSCognitoIdentityInteractiveAuthenticationDelegate {
func startNewPasswordRequired() -> AWSCognitoIdentityNewPasswordRequired {
//assume we are presenting from login vc cuz where else would we be presenting that from
DispatchQueue.main.async {
let presentVC = UIApplication.shared.keyWindow?.visibleViewController
TransitionHelperFunctions.presentResetPasswordViewController(viewController: presentVC!)
print(1)
}
var vcToReturn: ResetPasswordViewController?
returnVC { (vc) in
vcToReturn = vc
print(2)
}
print(3)
return vcToReturn!
}
//put this into its own func so we can call it on main thread
func returnVC(completion: #escaping (ResetPasswordViewController) -> () ) {
DispatchQueue.main.sync {
let storyboard = UIStoryboard(name: "ResetPassword", bundle: nil)
let resetVC = storyboard.instantiateViewController(withIdentifier: "ResetPasswordViewController") as? ResetPasswordViewController
completion(resetVC!)
}
}
}
My relevant ResetPasswordViewController code is here:
class ResetPasswordViewController: UIViewController, UITextFieldDelegate {
#IBAction func resetButtonPressed(_ sender: Any) {
var userAttributes: [String:String] = [:]
userAttributes["given_name"] = firstNameField.text!
userAttributes["family_name"] = lastNameField.text!
let details = AWSCognitoIdentityNewPasswordRequiredDetails(proposedPassword: self.passwordTextField.text!, userAttributes: userAttributes)
self.newPasswordCompletion?.set(result: details)
}
}
extension ResetPasswordViewController: AWSCognitoIdentityNewPasswordRequired {
func getNewPasswordDetails(_ newPasswordRequiredInput: AWSCognitoIdentityNewPasswordRequiredInput, newPasswordRequiredCompletionSource: AWSTaskCompletionSource<AWSCognitoIdentityNewPasswordRequiredDetails>) {
self.newPasswordCompletion = newPasswordRequiredCompletionSource
}
func didCompleteNewPasswordStepWithError(_ error: Error?) {
DispatchQueue.main.async {
if let error = error as? NSError {
print("error")
print(error)
} else {
// Handle success, in my case simply dismiss the view controller
SCLAlertViewHelperFunctions.displaySuccessAlertView(timeoutValue: 5.0, title: "Success", subTitle: "You can now login with your new passowrd", colorStyle: Constants.UIntColors.emeraldColor, colorTextButton: Constants.UIntColors.whiteColor)
self.dismiss(animated: true, completion: nil)
}
}
}
}
Thank you so much for your help in advance and let me know if you need any more information.

Where to trigger Loading in Clean Architeture Swift

Where is the correct place I should put the code that would trigger a loading to display in my app.
It is correct to do is on view? since it is displaying something on screen, so it fits as a UI logic
class ViewController: UIViewController {
func fetchData() {
showLoading()
interactor?.fetchData()
}
}
or on interactor? since it's a business logic. something like, everytime a request is made, we should display a loading. View only knows how to construct a loading, not when to display it.
class Interactor {
func fetchData() {
presenter?.presentLoading(true)
worker?.fetchData() { (data) [weak self] in
presenter?.presentLoading(false)
self?.presenter?.presentData(data)
}
}
}
same question applies to MVVM and MVP.
it is totally up to you . i am showing loading using an Observable .
in my viewModel there is an enum called action :
enum action {
case success(count:Int)
case deleteSuccess
case loading
case error
}
and an Observable of action type :
var actionsObservable = PublishSubject<action>()
then , before fetching data i call onNext method of actionObservable(loading)
and subscribing to it in viewController :
vm.actionsObserver
.observeOn(MainScheduler.instance)
.subscribe(onNext: { (action) in
switch action {
case .success(let count):
if(count == 0){
self.noItemLabel.isHidden = false
}
else{
self.noItemLabel.isHidden = true
}
self.refreshControl.endRefreshing()
self.removeSpinner()
case .loading:
self.showSpinner(onView : self.view)
case .error:
self.removeSpinner()
}
}, onError: { (e) in
print(e)
}).disposed(by: disposeBag)
You can use the delegate or completion handler to the update the UI from view model.
class PaymentViewController: UIViewController {
// for UI update
func showLoading() {
self.showLoader()
}
func stopLoading() {
self.removeLoader()
}
}
protocol PaymentOptionsDelegate : AnyObject {
func showLoading()
func stopLoading()
}
class PaymentOptionsViewModel {
weak var delegate : PaymentOptionsDelegate?
func fetchData() {
delegate?.showLoading()
delegate?.stopLoading()
}
}

Why RxSwift Subscribe just run once in First launch viewWillAppear?

I write a subscribe in viewWillAppear.
But it also run once in first launch app.
When I push to another viewcontroller, I use dispose().
Then I back in first viewcontroller, my subscribe func in viewWillAppear don't run.
What's wrong with my rx subscribe?
var listSubscribe:Disposable?
override func viewWillAppear(_ animated: Bool) {
super.viewWillAppear(animated)
listSubscribe = chatrooms.notifySubject.subscribe({ json in
print("*1") //just print once in first launch
self.loadContents()
})
}
override func viewWillDisappear(_ animated: Bool) {
super.viewWillDisappear(animated)
let controllers = tabBarController?.navigationController?.viewControllers
if (controllers?.count)! > 1 {
listSubscribe?.dispose()
}
}
RxSwift documentation says "Note that you usually do not want to manually call dispose; this is only an educational example. Calling dispose manually is usually a bad code smell."
Normally, you should be doing something like this -
let disposeBag = DisposeBag()
override func viewDidLoad() {
super.viewDidLoad()
whatever.subscribe(onNext: { event in
// do stuff
}).disposed(by: self.disposeBag)
}
As for your question, I believe you don't need to re-subscribe because you subscription will be alive and 'notifySubject' will send you updates whenever there are any.
Maybe you can get some reactive implementation of viewWillAppear and similar functions? And forget about manual disposables handling... For example your UIViewController init will contain something like this:
rx.driverViewState()
.asObservable()
.filter({ $0 == .willAppear })
.take(1) // if you need only first viewWillAppear call
.flatMapLatest({ _ in
// Do what you need
})
And the implementation of driverViewState:
public extension UIViewController {
public enum ViewState {
case unknown, didAppear, didDisappear, willAppear, willDisappear
}
}
public extension Reactive where Base: UIViewController {
private typealias _StateSelector = (Selector, UIViewController.ViewState)
private typealias _State = UIViewController.ViewState
private func observableAppearance(_ selector: Selector, state: _State) -> Observable<UIViewController.ViewState> {
return (base as UIViewController).rx
.methodInvoked(selector)
.map { _ in state }
}
func driverViewState() -> Driver<UIViewController.ViewState> {
let statesAndSelectors: [_StateSelector] = [
(#selector(UIViewController.viewDidAppear(_:)), .didAppear),
(#selector(UIViewController.viewDidDisappear(_:)), .didDisappear),
(#selector(UIViewController.viewWillAppear(_:)), .willAppear),
(#selector(UIViewController.viewWillDisappear(_:)), .willDisappear)
]
let observables = statesAndSelectors
.map({ observableAppearance($0.0, state: $0.1) })
return Observable
.from(observables)
.merge()
.asDriver(onErrorJustReturn: UIViewController.ViewState.unknown)
.startWith(UIViewController.ViewState.unknown)
.distinctUntilChanged()
}
}

how can I handle properly the behavior of camera in my Swift app?

In my swift app I'm allowing users to take photos.
For that purpose I've decided to use CameraManager from here: https://github.com/imaginary-cloud/CameraManager
When user opens my app, he sees a button - when he presses it, the camera view appears and he can take a photo. He also can dismiss the camera view and later on, at some point, press the button one more time to open camera view again.
If I understand it correctly from the plugin docs, I need to add a camera view to my view during first usage, then - in case of dismiss - invoke stopCaptureSession(), and during every next usage call resumeCaptureSession().
Currently in my swift code I have three methods:
let cameraManager = CameraManager()
fileprivate func addCameraToView()
{
cameraManager.addPreviewLayerToView(cameraView, newCameraOutputMode: CameraOutputMode.stillImage)
}
fileprivate func stopCaptureSession() {
cameraManager.stopCaptureSession()
}
fileprivate func resumeCaptreSession() {
cameraManager.resumeCaptureSession()
}
The IBAction for the button has the following code:
let currentCameraState = cameraManager.currentCameraStatus()
if currentCameraState == .notDetermined {
cameraManager.askUserForCameraPermission({ permissionGranted in
if permissionGranted {
self.resumeCaptreSession()
}
})
} else if (currentCameraState == .ready) {
self.resumeCaptreSession()
} else {
print("we do not have access to camera")
}
and in the IBAction for the dismiss button I had:
print("cancelling camera")
stopCaptureSession()
To make it work properly, I need to call addCameraToView() somewhere earlier - until now I was adding it in viewDidLoad, but I realized that I cannot do that because while doing so - the camera stays active until user presses the dismiss button.
So I thought about changing my code in IBAction for the camera button and add a camera from there. However, I have to add it only in case it wasn't add before - in the other case I need to call resumeCaptureSession().
The problem is that in CameraManager the function responsible for adding camera to the view is declared like this:
open func addPreviewLayerToView(_ view: UIView, newCameraOutputMode: CameraOutputMode) -> CameraState {
return addLayerPreviewToView(view, newCameraOutputMode: newCameraOutputMode, completion: nil)
}
open func addLayerPreviewToView(_ view: UIView, newCameraOutputMode: CameraOutputMode, completion: ((Void) -> Void)?) -> CameraState {
if _canLoadCamera() {
if let _ = embeddingView {
if let validPreviewLayer = previewLayer {
validPreviewLayer.removeFromSuperlayer()
}
}
if cameraIsSetup {
_addPreviewLayerToView(view)
cameraOutputMode = newCameraOutputMode
if let validCompletion = completion {
validCompletion()
}
} else {
_setupCamera({ Void -> Void in
self._addPreviewLayerToView(view)
self.cameraOutputMode = newCameraOutputMode
if let validCompletion = completion {
validCompletion()
}
})
}
}
return _checkIfCameraIsAvailable()
}
and resumeCaptureSession() is defined like this:
open func resumeCaptureSession() {
if let validCaptureSession = captureSession {
if !validCaptureSession.isRunning && cameraIsSetup {
validCaptureSession.startRunning()
_startFollowingDeviceOrientation()
}
} else {
if _canLoadCamera() {
if cameraIsSetup {
stopAndRemoveCaptureSession()
}
_setupCamera({Void -> Void in
if let validEmbeddingView = self.embeddingView {
self._addPreviewLayerToView(validEmbeddingView)
}
self._startFollowingDeviceOrientation()
})
}
}
}
So my question is - when user opens camera view, how can I check if camera was added to the view before, and if it was added - call resumeCaptureSession(), otherwise do not call it and just leave it with calling addCameraToView?

REST API, Swift, Automatic Update

I'm currently struggling to find an easy-to-use programming approach/design pattern, which solves the following problem:
I've got an REST API where the iOS app can request the required data. The data is needed in different ViewControllers. But the problem is, that the data should "always" be up to date. So I need to set up a timer which triggers a request every 5-20 seconds, or sth like that. Everytime the data changes, the view needs to be updated (at the current viewcontroller, which is displayed).
I tried some stuff with delegation and MVC Pattern, but it's kind a messy. How is it done the right way?
In my current implementation I only can update the whole UICollectionView, not some specific cells, because I don't know how the data changed. My controller keeps track of the data from the api and updates only if the hash has changed (if data changed on the server). My models always holds the last fetched data.
It's not the perfect solution, in my opinion..
I also thought about models, that keep themselves up to date, to abstract or virtualise my Rest-API. In this case, my controller doesn't even know, that it isn't directly accessible data.
Maybe someone can help me out with some kind of programming model, designpattern or anything else. I'm happy about anything!
UPDATE: current implementation
The Controller, which handles all the data
import Foundation
import SwiftyJSON
import SwiftyTimer
class OverviewController {
static let sharedInstance = OverviewController()
let interval = 5.seconds
var delegate : OverviewControllerUpdateable?
var model : OverviewModel?
var timer : NSTimer!
func startFetching() -> Void {
self.fetchData()
timer = NSTimer.new(every: interval) {
self.fetchData()
}
timer.start(modes: NSRunLoopCommonModes)
}
func stopFetching() -> Void {
timer.invalidate()
}
func getConnections() -> [Connection]? {
return model?.getConnections()
}
func getConnectionsSlave() -> [Connection]? {
return model?.getConnectionsSlave()
}
func getUser() -> User? {
return model?.getUser()
}
func countConnections() -> Int {
if let count = model?.getConnections().count {
return count
}
return 0
}
func countConnectionsSlave() -> Int {
if let count = model?.getConnectionsSlave().count {
return count
}
return 0
}
func fetchData() {
ApiCaller.doCall(OverviewRoute(), completionHandler: { (data, hash) in
if let actModel = self.model {
if (actModel.getHash() == hash) {
//no update required
return
}
}
var connections : [Connection] = []
var connectionsSlave : [Connection] = []
for (_,connection):(String, JSON) in data["connections"] {
let connectionObj = Connection(json: connection)
if (connectionObj.isMaster == true) {
connections.append(connectionObj)
} else {
connectionsSlave.append(connectionObj)
}
}
let user = User(json: data["user"])
//model needs update
let model = OverviewModel()
model.setUser(user)
model.setConnections(connections)
model.setConnectionsSlave(connectionsSlave)
model.setHash(hash)
self.model = model
//prevent unexpectedly found nil exception
if (self.delegate != nil) {
self.delegate!.reloadView()
}
}, errorHandler: { (errors) in
}) { (progress) in
}
}
}
protocol OverviewControllerUpdateable {
func reloadView()
}
The model, which holds the data:
class OverviewModel {
var user : User!
var connections : [Connection]!
var connectionsSlave : [Connection]!
var connectionRequests : [ConnectionRequest]!
var hash : String!
...
}
And in the ViewController, I use it like this:
class OverviewVC: UIViewController, UICollectionViewDataSource, UICollectionViewDelegate, OverviewControllerUpdateable {
let controller = OverviewController.sharedInstance
override func viewDidLoad() {
super.viewDidLoad()
self.controller.delegate = self
self.controller.startFetching()
}
//INSIDE THE UICOLLECTIONVIEW DELEGATE METHODS
...
if let user : User = controller.getUser() {
cell.intervalTime = interval
cell.nameLabel.text = "Ihr Profil"
}
...
func reloadView() {
self.userCollectionView.reloadData()
}
}
You could use a Singleton object to fetch your data periodically, then post notifications (using NSNotificationCenter) when the data is updated. Each view controller dependent on the data would listen for these notifications, then reload UI based on the updated data.

Resources