Running custom method of the main thread - ios

class func loadData(
onCompletition: #escaping ([LocationInfo])->Void){
let workingQueue = DispatchQueue.global(qos:.utility)
let completitionQueue = DispatchQueue.main
workingQueue.sync {
print("\n Data fetch started \n")
let root = FIRDatabase.database().reference()
let locationSummary = root.child("LocSummary")
locationSummary.observeSingleEvent(of: .value,with: { (snapshot) in
for item in snapshot.children{
let locationInfo = LocationInfo(snapshot: item as! FIRDataSnapshot)
FirebaseDataController.resultsArray.append(locationInfo)
}
completitionQueue.async {
print("\n data fetch completed \n ")
onCompletition(FirebaseDataController.resultsArray)
print("After on completion method")
}
})
}
}
The problem I have no is that, every time I want to access the data inside the results array I have to go through this functions completion handler. Which is not something I can do all the time specially when I want to work with table views and such (I have a seperate class to handle all DB interactions and many other classes to handle table view interactions).
My objective is to run this code at the start of the application may be through the AppDelegate and have a populated array that I can call anytime I want access to data.
To do this I think I need to run this code on the main thread. I tried that by substituting the workingQueue with the main thread but the application keeps crashing.
Is there is anything that I can do about this?

Your application will crash only if you call this synchronously on main thread when launching application because you cause a deadlock. If you run this function in application:didFinishLaunching: method of AppDelegate you should be able to use safely this implementation:
class func loadData(onCompletition: #escaping ([LocationInfo])->Void){
let completitionQueue = DispatchQueue.main
print("\n Data fetch started \n")
let root = FIRDatabase.database().reference()
let locationSummary = root.child("LocSummary")
locationSummary.observeSingleEvent(of: .value,with: { (snapshot) in
for item in snapshot.children{
let locationInfo = LocationInfo(snapshot: item as! FIRDataSnapshot)
FirebaseDataController.resultsArray.append(locationInfo)
}
completitionQueue.async {
print("\n data fetch completed \n ")
onCompletition(FirebaseDataController.resultsArray)
print("After on completion method")
}
})
}
Your guess that you need to perform this operations on the main thread is wrong - there is no reason to do that, in fact, you will end up with stalled user interface. Fetch the data, and then asynchronously update user interface on main thread.

Since you want to call the method on main thread.
First thing you remove all the code related to DispatchQueue.
Then call the loadData using - (void)performSelectorOnMainThread:(SEL)aSelector withObject:(id)arg waitUntilDone:(BOOL)wait;

Related

CoreData multithreading fetch requires small delay before accessing attributes

I use coreData as persistent store.
To read data, I use (only essential parts are shown):
func fetchShoppingItems(completion: #escaping (Set<ShoppingItem>?, Error?) -> Void) {
persistentContainer.performBackgroundTask { (backgroundManagedContext) in
let fetchRequest: NSFetchRequest<CDShoppingItem> = CDShoppingItem.fetchRequest()
var shoppingItems: Set<ShoppingItem> = []
do {
let cdShoppingItems: [CDShoppingItem] = try backgroundManagedContext.fetch(fetchRequest)
for nextCdShoppingItem in cdShoppingItems {
nextCdShoppingItem.managedObjectContext!.performAndWait {
Thread.sleep(forTimeInterval: 0.1) // This seems to be required
let nextShoppingItem = ShoppingItem.init(name: nextCdShoppingItem.name!)
shoppingItems.insert(nextShoppingItem)
} // performAndWait
} // for all cdShoppingItems
completion(shoppingItems, nil)
return
} catch let error as NSError {
completion(nil, error)
return
} // fetch error
} // performBackgroundTask
} // fetchShoppingItems
To test the coreData implementation, I wrote a unit test that creates multiple threads that write to and read from coreData concurrently.
This test runs only successfully, if the instruction
Thread.sleep(forTimeInterval: 0.1) // This seems to be required
is inserted in the performAndWait closure.
If it is commented out, nextCdShoppingItem is often read back with nil attributes, and the function crashes due to the forced unwrap.
I am not sure, if nextCdShoppingItem.managedObjectContext!.performAndWait is correct or if I had to use backgroundManagedContext.performAndWait, but with backgroundManagedContext the effect is the same.
I do not understand why inserting a small delay before accessing an attribute of a managed object is necessary to avoid the problem.
Any hint is welcome!
EDIT:
I investigated the issue further, and found the following:
Every time nextCdShoppingItem is read back by the background thread (called read thread below) as nil, there is also another background thread that tries to save its own managedContext after all records in its managedContext have been deleted (called write thread below).
Apparently the read thread tries to fetch a record that has just been deleted by the write thread.
So the problem is definitively a multithreading issue, and I found a solution (see my answer below).
performAndWait will add the block to the queue and schedule it to run, just like perform, but performAndWait will not return until the block is complete. Since you are inside a loop of cdShoppingItems, the loop does not stop and wait for the block to return. By adding the thread sleep, you are essentially slowing down the loop and giving core data enough time to complete its fetch. The forced unwrap crash is probably an indication that it's lost its nextCdShoppingItem reference.
I would consider refactoring where you do not need to query core data inside a loop. If it's possible, add the name attribute to CDShoppingItem so you don't have to fetch it to build a ShoppingItem object.
Edit: took a stab at a refactor although I don't know your exact use case:
func fetchShoppingItems(completion: #escaping (Set<ShoppingItem>?, Error?) -> Void) {
persistentContainer.performBackgroundTask { (backgroundManagedContext) in
let fetchRequest: NSFetchRequest<CDShoppingItem> = CDShoppingItem.fetchRequest()
do {
var shoppingItems: Set<ShoppingItem> = []
let cdShoppingItems: [CDShoppingItem] = try backgroundManagedContext.fetch(fetchRequest)
for nextCdShoppingItem in cdShoppingItems {
if let name = nextCdShoppingItem.name {
let nextShoppingItem = ShoppingItem.init(name: name)
shoppingItems.insert(nextShoppingItem)
}
}
completion(shoppingItems, nil)
} catch let error as NSError {
print("Error fetching CDShoppingItem: \(error)")
completion(nil, error)
} // fetch error
return
} // performBackgroundTask
} // fetchShoppingItems
To prevent the multithreading issue, I tried 2 things:
1) Since iOS10, a persistentStore of SQL type maintains a connection pool for concurrent access to the pool, and it is possible to set a maximum pool size, see the WWDC video. I did so using
private lazy var persistentContainer: NSPersistentContainer = {
let container = NSPersistentContainer(name: modelName)
container.loadPersistentStores(completionHandler: { (storeDescription, error) in
if let error = error as NSError? {
// ...
} else {
storeDescription.setOption(NSNumber(1), forKey: NSPersistentStoreConnectionPoolMaxSizeKey)
}
})
return container
}()
to prevent concurrent access to the persistentStore. Unfortunately, this did not solve the problem for an unknown reason.
2) I then tried to serialize read and write operations by setting up a serial queue:
private let coreDataSerialQueue = DispatchQueue(label: "com.xxx.ShopEasy.coreDataManager") // serial by default
It is used for the read and write accesses in the following way:
coreDataSerialQueue.async {
let backgroundManagedContext = self.persistentContainer.newBackgroundContext()
backgroundManagedContext.performAndWait {
// …
} // performAndWait
} // coreDataSerialQueue.async
This did solve the problem.
Please note that it would be wrong to use
coreDataSerialQueue.async {
self.persistentContainer.performBackgroundTask { (backgroundManagedContext) in
// …
} // performBackgroundTask
} // coreDataSerialQueue.async
because performBackgroundTask would fork another asynchronous thread and thus break the serialization.

How do I ensure that all network calls have been made before accessing my core data model?

I am making multiple api calls in succession and when I finally push to my next view controller my data comes up completely blank from my core data model. In ViewController A I have made the following requests in this order:
Api.verifyOtp(email, otp).continueWith { (task) -> Any? in
if task.succeed {
self.apiCallOne()
self.apiCallTwo()
self.apiCallThree()
self.apiCallFour()
self.apiCallFive()
} else {
Hud.hide()
task.showError()
}
return nil
}
Now all of these calls are made asynchronously. However the last method which is self.apiCallFive() is the method that pushes to ViewController B. Here is the call:
Api.apiCallFive().continueOnSuccessWith { (task) -> Any? in
Hud.hide()
if task.succeed {
let storyboard = UIStoryboard(name: "Main", bundle: nil)
let viewB storyboard.instantiateViewController(withIdentifier: "ViewControllerB" self.navigationController?.pushViewController(viewB, animated: true)
}
My guess is that since all of these requests are happening asynchronously then there's no guarantee on which call will finish first. So the apiCallFive() is pushing and loading ViewController B before the others are able to finish. How can I make it so the next view will not be loaded or pushed to until all of the tasks have been completed?
Thank you!
I have faced the same issue. Fix it by using DispatchGroup.
Code:
Define as property
let APIGroup = DispatchGroup()
Execute below code when any API Calling starts.
APIGroup.enter()
Execute below code when any API Calling Completed.
downloadGroup.leave()
Notify Block:
APIGroup.notify(queue: DispatchQueue.main) {
print("All APIs called successfully: Perform required operation")
}
There no need to manage by any counter or other variables. notify block call automatically when all task completed successfully.
What’s really important here is the enter-leave pairs. You have to be
very careful and make sure that you leave the group. It would be easy
to introduce a bug in the code above. Let’s say that we didn’t leave
the group in that guard statement above, just before the return. If
the API called failed, or the JSON was malformed, the number of groups> entries would not match the number of leaves. So the group completion
handler would never get called. If you’re calling this method from the
UI and displaying an activity indicator while your networking requests
are running, you would never get a callback from the method, and you
would keep on spinning 🙂
Apple documents
To solve this you need a way of getting notified when each call is finished.
The easiest way of doing this is using completion blocks on each call.
func apiCall(completion: #escaping () -> Void) {
....
}
After adding completion blocks to the api calls, your blocks could look like this:
let dispatchGroup = DispatchGroup()
dispatchGroup.enter()
apiCallOne {
dispatchGroup.leave()
}
dispatchGroup.enter()
apiCallTwo {
dispatchGroup.leave()
}
...
dispatchGroup.enter()
apiCallN {
dispatchGroup.leave()
}
dispatchGroup.wait(timeout: Constants.timeout)
Keep in mind that the wait statement will block the thread where you call it until all the leave() statements are executed, so be careful that you don't end up with a deadlock.

Cancel UICollectionView updates when view is being dismissed

I have recently reported a crash in my app, and I've found out what is happening and I need some help/best practices/best approach to this issue.
I have a pushed UICollectionViewController that on viewDidLoad queries the server to fetch some data to fill the UICollectionView.
My problem here is, if I push this UICollectionViewController and then tap the back button fast - the background thread still continues to fetch the server data, but when the data is fetched I update the UICollectionView with the performBatchUpdates() and my app crashes.
Here it happens because the app is attempting to reload data on a view that's not visible anymore.
What's the best practice here?
Is there any way to "abort" collection view updates if I'm moving back to the previous VC?
something like:
if self.isMovingFromParentViewController { /* abort any update here? */ }
Thanks
You can use DispatchWorkItem for achieving this as follows
let backgroundQueue = DispatchQueue.global()
var backgroundTask: DispatchWorkItem!
backgroundTask = DispatchWorkItem { [weak self] in
// Perform background task
if !backgroundTask.isCancelled {
return to main Queue
}
backgroundTask = nil // resolve strong reference cycle
}
backgroundQueue.async(execute: backgroundTask)
// When you want to cancel the task
backgroundQueue.async { [weak backgroundTask] in
backgroundTask?.cancel()
}
This is desirable in many cases where we should abort all the Server request. I prefer to perform all the clean up in the
deinit() {
// Abort all your APIs and asynchronous call
// Release all dependency
}
Along with this, always have a weak reference of your controllers and then perform optional binding in the response of the Asynchronous call.
Almofire.request(reqData: param, method: get.....) {
[weak self] response in
guard let safeSelfRef = self, let safeCollectionView =
safeSelfRef.collectionView else { return }
//Update view here
}

Cannot signal semaphore with Firebase observeEventType - what gives?

I am using Firebase to read/write data to a remote server.
My objective is to write a routine that also takes a callback function. Heres what the routine should accomplish:
Write some data to Firebase (this works fine)
Observe for changes on above written data
If change occurs call callback with success
If a timeout is triggered, call callback with failure
Im unable to figure out why steps 2-4 are failing for me. Below is a snippet of the code
Here's whats going on in the code:
It sets up a GCD semaphore
It writes some data to the root firebase URL - specifically it adds a user called joe. This works fine. I can see the data written from Vulcan
it sets up a reference to Joe's node and watches for events on childAdded for Joe. Here is the API reference
Upon successfully getting an event it signals the semaphore, which should call the callback with message "Success"
Instead I see the "timeout" message. Strangely though the observer does print out the newly added key:value pair, which is doubly confusing!
Whats going on here? Am I using semaphores incorrectly? Or am i using Firebase's API incorrectly?
typealias completionfoo = (gid:String) -> Void
func setStateReady(somestr: String, callfoo: completionfoo) {
let semaphore = dispatch_semaphore_create(0)
var myRootRef = Firebase(url:"https://xxxxxx.firebaseio.com/")
var joe = ["state" : "READY", "msg" : somestr]
var usersRef = myRootRef.childByAppendingPath("users")
var users = ["joe": joe]
usersRef.setValue(users)
var joeref = Firebase(url: "https://xxxxxx.firebaseio.com/users/joe/")
// Read data and react to changes
joeref.observeEventType(.ChildAdded, withBlock: {
snapshot in
print(snapshot.key)
print(snapshot.value)
dispatch_semaphore_signal(semaphore)
})
let timeout = dispatch_time(DISPATCH_TIME_NOW, Int64(30 * Double(NSEC_PER_SEC)))
if dispatch_semaphore_wait(semaphore, timeout) != 0 {
joeref.removeAllObservers()
callfoo(gid: "timeout")
} else {
joeref.removeAllObservers()
callfoo(gid: "success")
}
}
override func viewDidLoad() {
super.viewDidLoad()
setStateReady("hi i am joe") { (gid) -> Void in
print(gid)
}

Best way to use findObjectsInBackgroundWithBlock when moving data around

I have a Tableview that gets data with findObjectsInBackgroundWithBlock in viewDidLoad and passes that data to a Detail View Controller no problem.
Im having trouble managing the flow of findObjectsInBackgroundWithBlock. Here is a example: I have a like button on the detail view and when pressed it increments the UILabel and displays it. It also then gets the object in Parse then increments and saves it... Everything good.
#IBAction func likeButtonPressed(sender: AnyObject) {
print("likeButtonPressed()")
// Adding the like to label
mixLike!++
var stringForCount: String = String(mixLike!)
mixLikeLabel.text = stringForCount
// Saving the like back to Parse
var query = PFQuery(className: "musicMixes")
query.whereKey("info", equalTo: mixNameLabel.text)
query.findObjectsInBackgroundWithBlock { (objects:[AnyObject]!, error:NSError!) -> Void in
if error == nil {
for object in objects {
//var votes = object["votes"] as! Int
let mixObject:PFObject = object as! PFObject
mixObject.incrementKey("votes", byAmount: 1)
mixObject.saveInBackgroundWithTarget(nil, selector: nil)
print("mixObjectSaved")
}
} else {
print("Error getLikeCount()")
}
print("sending Notification...")
NSNotificationCenter.defaultCenter().postNotificationName("reload", object: nil)
print("sent Notification...")
}
} // likeButtonPressed End
I also then call a NSNotification back to the Table View so the Table View can update the likes to match the users like click on the detail view (See bellow)
The NSNotification calls this function in the Table View, which removes the like array, grabs the new likes again and then reloads the Table View.
# objc func reloadTableData(notification: NSNotification){
print("Notification Recived, Removing Likes and Reloading. reloadTableData()...")
self.mixLikeArray.removeAll()
//self.stringForCountArray.removeAll()
print("Like array Data removed, getting data again...")
var query = PFQuery(className: "musicMixes")
query.orderByAscending("date")
query.findObjectsInBackgroundWithBlock { (objects:[AnyObject]!,error: NSError!) -> Void in
if error == nil {
for object in objects {
let mixLike = object["votes"] as! Int
self.mixLikeArray.append(mixLike)
print("New mixLikeArray data is \(self.mixLikeArray)")
}
} else {
print("error getting like object")
}
}
dispatch_async(dispatch_get_main_queue(),{
self.allTableView.reloadData()
});
}
I see three issues wrong with how this works at the moment. likeButtonPressed() Is sometimes sending the NSNotification before mixObject.saveInBackgroundWithTarget is finished. Meaning that the incremented like won't be displayed on the table view.
Secondly if I was to click like then click back to tableview swiftly the app will crash. This is because I'm guessing both likeButtonPressed() and the NSNotification function still has not been completed.
Also in # objc func reloadTableData(notification: NSNotification) once again the
dispatch_async(dispatch_get_main_queue(),{
self.allTableView.reloadData()
});
Is being called before the findObjectsInBackgroundWithBlock is being completed? Anyway round this?
How would you suggest I can remodel this to work efficiently? Im pretty new to coding and a bit rusty with designing the best ways to do things... I know the concept behind completion handlers could I use these? I know that Parse likes to work in the background though hhhmmmm.....
to fix your reloadTableData problem, you should trigger the reload once the parse block is done executing, which means moving this line
dispatch_async(dispatch_get_main_queue(),{
self.allTableView.reloadData()
});
inside the block
query.findObjectsInBackgroundWithBlock { (objects:[AnyObject]!,error: NSError!) -> Void in
if error == nil {
for object in objects {
let mixLike = object["votes"] as! Int
self.mixLikeArray.append(mixLike)
print("New mixLikeArray data is \(self.mixLikeArray)")
}
dispatch_async(dispatch_get_main_queue(),{
self.allTableView.reloadData()
});
} else {
print("error getting like object")
}
}
That will ensure that it gets triggered once parse is done updating objects. Currently its triggering before that while the block is executing. It also means that it won't reload if you get an error as you probably need to handle that differently anyway.
As for your problem of the notification happening before the saving is complete, you are calling . saveInBackgroundWithTarget but don't seem to send anything into it. You could use saveInBackgroundWithBlock and then use dispatch_group dispatch_group_enter, dispatch_group_leave, and dispatch_group_notify inside the block to make your program wait till everything is done being saved before sending the notification.
So you would create a dispatch_group
dispatch_group_t group = dispatch_group_create();
And then call it dispatch_group_enter in the for loop through the objects
for object in objects {
dispatch_group_enter(group);
let mixObject:PFObject = object as! PFObject
.....
}
Then call dispatch_group_leave on the mixObject.saveInBackgroundWithBlock
and wrap the notification in dispatch_group_notify
dispatch_group_notify(group, dispatch_get_main_queue(), ^{ // 4
NSNotificationCenter.defaultCenter().postNotificationName("reload", object: nil)
});
Something like that
It sounds more daunting than it is, here's a Ray Wenderlich tutorial to bring you up to speed on how to use it, if your not familiar

Resources