In one of my apps users have an ability to search for a location by its name. I am using Google Maps APIs to display autocompleted suggestions with the help of GMSPlacesClient.
Code below illustrates how I do this:
func performSearch() {
let filter = GMSAutocompleteFilter()
filter.type = GMSPlacesAutocompleteTypeFilter.City
placesClient?.autocompleteQuery(searchBar.text!, bounds: nil, filter: filter, callback: { (results, error: NSError?) -> Void in
// update results
})
}
This code works very well. However, I noticed one scenario in which it sort of fails. One example is if I search for "Chicago". If I type "Ch" very fast sometimes the results are returned for "Ch" correctly, but sometimes I obtain the results for "C". The problem arises because these are asynchronous requests and I am performing the search each time user types something. So, even though the request for "C" was initiated before "Ch" it can return last.
I therefore need to cancel all previous initiated requests before I start a new one. However, I couldn't find a way of doing this. Does anyone know how to achieve it?
Update
I have tried to use a workaround:
let string = myResults[0].attributedFullText
string.enumerateAttribute(kGMSAutocompleteMatchAttribute, inRange: NSMakeRange(0, string.length), options: .Reverse) { (value, range, stop) -> Void in
if value != nil {
let t = NSString(string: string.string)
let str = t.substringWithRange(range)
//now compare with original search string
}
}
This fixes the issue with Chicago (and other similar cases). However, such approach will disable autocomplete suggestions when user makes a typo because the matched string and original search string are always different in this case.
So, the question still remains open. An ideal approach would be to cancel all previously initiated requests. If Google also stored the initial search string an approach similar to that presented above could work, but it looks like the original search string is not returned after request...
It seems like the only option is to not use the SDK but to use Alamofire to send requests. Those requests can then be easily cancelled. However, I am not sure how good this is...
Related
My project consist of obj-c and swift classes. I use Firebase 7.3.0.
I manually log screen_view event for my screens. I call this method in viewWillAppear or viewDidAppear like this:
#objc class MyAnalyticConstants: NSObject {
static let myScreenName = "AwesomeScreen"
}
class MyViewController: UIViewController {
override func viewWillAppear(_ animated: Bool) {
super.viewWillAppear(animated)
MyAnalyticsClass().logScreenViewEvent(name: MyAnalyticConstants.myScreenName)
}
}
class MyAnalyticsClass {
func logScreenViewEvent(name: String, parameters: [String: Any]? = nil) {
var param = [String: Any]()
if let parameters = parameters {
param = parameters
}
param[AnalyticsParameterScreenName] = name
logEvent(name: AnalyticsEventScreenView, parameters: param)
}
func logEvent(name: String, parameters: [String: Any]?) {
analytics.logEvent(name, parameters: parameters)
}
}
I turned off automatic screenview reporting by setting FirebaseAutomaticScreenReportingEnabled to NO (Boolean) in the Info.plist. I use struct with static names for my screens.
However, sometimes I see (not set) value for "screen_view" event inside my google analytics path exploration for production. I can't catch this while using DebugView.
screenshot
I would really appreciate it, if somebody could help me to fix or explain it.
EDIT:
I swizzled firebase method
+ (void)xxx_logEventWithName:(NSString *)name
parameters:(nullable NSDictionary<NSString *, id> *)parameters {
[self xxx_logEventWithName:name parameters:parameters];
if ([name isEqualToString:#"screen_view"] && [parameters[#"screen_name"] length] <= 3) {
NSLog(#"%#", #[][1]);
}
and jumped through app during 30 min. I didn't catch up crash. Any other ideas?
Ok, there are a few things you could typically do in this case.
I wrote it as an afterthought, but it's something you should make sure of before doing technical debugging that follows: you should go to your analytics property/view and debug the filters. Maybe you have replacing filters interfering with your values, but I presume you checked your data in a full and unfiltered view where your app is the sole "stream" of data. This is important. The bug can come from a different app to this property, so either make sure you're the only source, or make sure you filter out other sources/apps/platforms.
Check your logEvent function calls where you either send AnalyticsEventScreenView or just the "screen_view" string as the first parameter. You see how the Firebase lib uses one function to send all kinds of events? They now treat screenviews as events. Which has its elegancy, but also may lead to unintended mistakes. Check what the globals actually mean in here: https://github.com/firebase/firebase-cpp-sdk/blob/0c8c8b29bc2d62d66c6ac49ff2c3fb04f815a687/analytics/ios_headers/FIREventNames.h
Check your logScreenViewEvent function calls. Pay attention to cases when you pass the first parameter as a variable. Also make sure you're never setting the AnalyticsParameterScreenName, which is also known as a string "screen_name" from here: https://github.com/firebase/firebase-cpp-sdk/blob/0c8c8b29bc2d62d66c6ac49ff2c3fb04f815a687/analytics/ios_headers/FIRParameterNames.h in the parameters dictionary, the second attribute. Cuz setting it there will effectively overwrite whatever is the first attribute you're setting. I actually usually suggest having only one argument for the screenview function declaration, especially to avoid collisions like this.
Oh, almost forgot. Make sure you ALWAYS use your neat MyAnalyticsClass wrapper and never call the native logEvent(). I would just check all files where you include the Firebase sdk and see if it should be replaces with the wrapper.
Finally, if the above didn't help, you can insert the check in both your function wrappers to throw an error whenever the event name equals to "screen_view" and the "screen_name" parameter's length not more than 2 character (I'm just trying to include all falsy values, so null, undefined, nil, whatever). And run unit tests or even better - regression testing with things set like that. Well, or manually test it out, watching for the errors in the console rather than the web debugger.
So my goal is to get rid of these bugs completely. I am in a dilemma where each decision leads to a bug.
The first thing I can do that eventually becomes an issue is use a String-interpolated collection path in all my query functions like so:
func getEventName() {
listener = db.collection("school_users/\(user?.uid)/events").order(by: "time_created", descending: true).addSnapshotListener(includeMetadataChanges: true) { (querySnapshot, error) in
if let error = error {
print("There was an error fetching the data: \(error)")
} else {
self.events = querySnapshot!.documents.map { document in
return EventName(eventName: (document.get("event_name") as! String))
}
self.tableView.reloadData()
}
}
}
The thing with this is, when I run the app on the simulator, I am restricted from pressing buttons and then sometimes I can press them and then sometimes they get restricted again. This bug is so confusing because it makes no sense where it springs from.
The other issue is I can use a Constants value in all the query functions in my collections path.
static let schoolCollectionName = "school_users/\(user?.uid)/events"
This is nested in a Firebase struct within the Constants struct. In order to keep Xcode from giving errors I create a let users = Auth.auth().currentUser variable outside the Constants struct. The issue with this value is that when I put that in all of my query functions collection paths, all the buttons are accessible and selectable all the time, but when a user logs out and I log in as a new user, the previous user's data shows up in the new user's tableview.
It would obviously make more sense to use the Constants value because you prevent typos in the future, but I can't figure out how to get rid of the bug where the old user's data shows up in the new user's tableview. Thanks in advance.
The user id should definitely not be a constant. What it sounds like is that right now, you have no reliable way to change users -- your setup probably depends on which user is logged in at app startup, since that's where your variable gets set.
I would do something more like this:
func getEventName() {
guard let user = Auth.auth().currentUser else {
//handle the fact that you don't have a user here -- don't go on to the next query
return
}
listener = db.collection("school_users/\(user.uid)/events").order(by: "time_created", descending: true).addSnapshotListener(includeMetadataChanges: true) { (querySnapshot, error) in
Note that now, user.uid in the interpolated path doesn't have the ? for optionally unwrapping it (which Xcode is giving you a warning for right now). It will also guarantee that the correct query is always made with the currently-logged-in user.
Regarding being able to press the buttons, that sounds like an unrelated issue. You could run your app in Instruments and check the Time Profiler to see if you have long-running tasks that are gumming up the main/UI thread.
I am sending a query of geometry to show the features selected on map and get the selected features.
Both the things are working okay but when i check the attribute dictionary of a feature it contains only 5 key/value pair but the same function in android returning 10 key/value pair.
I am making query like this
let query = AGSQueryParameters()
if let selectionGraphicGeometry = selectionGraphic?.geometry {
let geometry = AGSGeometryEngine.simplifyGeometry(selectionGraphicGeometry)
query.geometry = geometry
}
selectableLayer?.selectFeatures(withQuery: query, mode: AGSSelectionMode.add, completion: { (result, error) in
if let features = result?.featureEnumerator().allObjects {
for feature in features {
let keys = feature.attributes.allKeys
}
}
}
I dont know where i am doing this wrong
In the Version 100 Runtime, we take a slightly different approach for efficiency's sake.
Features will by default only include the minimal set of fields required for rendering and editing. When you are making a selection, you are working with those features, so that's why you're seeing the smaller set of fields.
If you need all the fields for your selected features, you should actually perform a query on your AGSServiceFeatureTable and select the features based off that.
Something like this:
let table = selectableLayer.featureTable as? AGSServiceFeatureTable
table?.queryFeatures(with: query, queryFeatureFields: .loadAll) { (result, error) in
guard error == nil else {
print("Error selecting features: \(error!.localizedDescription)")
return
}
guard let features = result?.featureEnumerator().allObjects else {
return
}
selectableLayer.select(features)
for feature in features {
let keys = feature.attributes.allKeys
print(keys)
}
}
What's odd is that you say you're seeing a different number of fields returned on Android than on iOS. Are you sure the Android app is displaying the same layer with the same renderer?
One other point: You might be better off using the Esri Community (GeoNet). The ArcGIS Runtime SDK for iOS Forum can be found here. Do post a question there if you are seeing different numbers of fields on iOS and Android with the same layer and renderer.
Hope this helps!
P.S. There are two related things you might want to know.
AGSArcGISFeature instances are now Loadable. So if you have an individual feature and you want to get all the fields for it from the source service, you can call load(completion:) on it. Or you could pass an [AGSArcGISFeature] array to the ASGLoadObjects() helper function. However, that function will make a separate network request for each feature so if your array isn't small, that could lead to a bad user experience.
You can set your AGSServiceFeatureTable.featureRequestMode to .manualCache. You then need to call populateFromServiceWithParameters() to load the precise data that you need locally (and as you pan and zoom the map, you will need to manually manage this cache). See here for more details.
1. Clicking the link causes a database update.
There is a certain link I have access to (let's pretend it's www.google.com), such that when I open it up in my browser, it updates a certain section of the JSON code in my database. Based on the numbers that make up a portion of the link, it adjusts a certain value in the data.
2. How do I run this link in the background of my iOS app?
I need to be able to "open" this link within the app, without actually opening up a UIWebview and visually visiting the site. I just need this JSON data inside the database to update on its own (with the user unaware that it even happened).
The problem I'm having here is that I simply don't know how this is done. How do I cause this link to be visited without opening up a Safari browser?
The best approach I've found for such functions is to treat them as if they were "AJAX" (or "REST", "API", etc.) - while these terms are often used (and for more seasoned programmers instantly give a certain thought), the end result is that they take information from your 'originator' and send to the 'server' for processing, which then replies with a 'response'. Once you get that concept in your head, this becomes a fairly simple activity.
(for our example, I will call this "API", as that does really suit {as #Mortiz suggested} this question best)
For Swift 3, there are several ways to do this, I'll show you two I've found and use for various functions:
DispatchQueue
For a 'one-time shot to a url that I know exists and will connect reliability', this is a good one to use (think of it as a 'quick-n-dirty' if you like....!)
DispatchQueue.global().async {
let data = try? Data(contentsOf: theURL!) //make sure your url does exist, otherwise unwrap in a if let check / try-catch
DispatchQueue.main.async {
// do stuff here with the data if you need
// you can get the response from the server and parse it out, set buttons in the app, etc.
}
}
Alamofire
For Swift 3, Alamofire is extremely popular and does a lot of great stuff. Check it out if you haven't already!
Alamofire.request("\(theURL!)").responseJSON { response in
print("result is ", response.result)
switch response.result {
case .success(let value):
// do stuff with the returned data
// like updating your internal database, etc.
print(value)
case .failure(let error):
print("There was an error")
// you can see the error response in various ways....
print(requested)
print(error)
print(response)
print(response.result)
}
}
Once you have your buttons in place (from your description it sounds like that is what your #1 is about), then in the function you call when it is clicked, drop in the code from above and 'do stuff' as you need.
This will make the update to the server automatically in the background (answering your #2) - the user won't notice anything unless there are connection issues to the internet, etc. (much too complex to get into here, though if you expect to have much of it, Alamofire is a great choice as it automatically retries, etc. (part of the great features you should check out)
A key piece of this is that you can take the response from the URL (send various bits of JSON data back from the server, then break it back down in the phone) and do 'whatever' with it.
Some things you can do (to hopefully give you more ideas - which is just about anything.....):
update data in the app (local storage, local variables, etc.)
update text (color, background) inside Buttons or Labels
process Alerts to the user (not your case, but sometimes you want to let them know what went on - certainly if it was an error in updating your server)
change Images (various things)
switch Views
Well, the list is as long as "things you can do in an app", so decide for yourself what you need to mod/update - this is "the" way to do it!
You could also use the UIWebView without ever showing it, like this (Swift 3):
func webView() {
let theWebView: UIWebView
theWebView = UIWebView(frame: UIScreen.main.bounds)
theWebView.delegate = self
if let theURL = URL(string: "your URL") {
let request = URLRequest(url: theURL)
theWebView.loadRequest(request)
}
}
Just don't add it to the view.
I was hoping that someone can help a coding newbie with what might be considered a stupid question. I'm making a blog type app for a community organization and it's pretty basic. It'll have tabs where each tab may be weekly updates, a table view with past updates and a tab with general information.
I setup cloudkit to store strings and pictures, and then created a fetchData method to query cloud kit. In terms of the code (sample below) it works and gets the data/picture. My problem is that it takes almost 5-10 seconds before the text and image update when I run the app. I'm wondering if that's normal, and I should just add an activity overlay for 10 seconds, or is there a way to decrease the time it takes to update.
override func viewDidLoad() {
fetchUpcoming()
}
func fetchUpcoming() {
let container = CKContainer.defaultContainer()
let publicData = container.publicCloudDatabase
let query = CKQuery(recordType: "Upcoming", predicate: NSPredicate(format: "TRUEPREDICATE", argumentArray: nil))
publicData.performQuery(query, inZoneWithID: nil) { results, error in
if error == nil { // There is no error
println(results)
for entry in results {
self.articleTitle.text = entry["Title"] as? String
self.articleBody.text = entry["Description"] as? String
let imageAsset: CKAsset = entry["CoverPhoto"] as! CKAsset
self.articlePicture.image = UIImage(contentsOfFile: imageAsset.fileURL.path!)
self.articleBody.sizeToFit()
self.articleBody.textAlignment = NSTextAlignment.Justified
self.articleTitle.adjustsFontSizeToFitWidth = true
}
}
else {
println(error)
}
}
}
Another question I had is about string content being stored on cloud kit. If I want to add multiple paragraphs to a blood entry (for example), is there a way to put it in one record, or do I have to separate the blog entry content into separate paragraphs? I may be mistaken but it seems like CloudKit records don't recognize line breaks. If you can help answer my questions, I'd be really appreciative.
It looks like you might be issuing a query after creating the data, which isn't necessary. When you save data, as soon as your completion block succeeds (with no errors) then you can be sure the data is stored on the server and you can go ahead and render it to the user.
For example, let's say you're using a CKModifyRecordsOperation to save the data and you assign a block of code to the modifyRecordsCompletionBlock property. As soon as that block runs and no errors are passed in, then you can render your data and images to your user. You have the data (strings, images, etc.) locally because you just sent them to the server, so there's no need to go request them again.
This provides a quicker experience for the user and reduces the amount of network requests and battery you're using on their device.
If you are just issuing normal queries when your app boots up, then that amount of time does seem long but there can be a lot of factors: your local network, the size of the image you're downloading, etc. so it's hard to say without more information.
Regarding the storage of paragraphs of text, you should consider using a CKAsset. Here is a quote from the CKRecord's documentation about string data:
Use strings to store relatively small amounts of text. Although
strings themselves can be any length, you should use an asset to store
large amounts of text.
You'll need to make sure you're properly storing and rendering line break characters between the user input and what you send to CloudKit.