I have a swiftUI view that calls a UIViewRepresentable view. In the SwiftUI view I am toggling the state of #State boolean value.
In my UIViewRepresentable view I have created a binding that gets past from the main SwiftUI view. The problem is the binding never gets update or at least the updateView function is not getting called in the UIViewRepresentable view. I fell like I must be doing something wrong but I am just overlooking it. Here is an example of what I am trying to do.
import Foundation
import SwiftUI
struct BindingTest: UIViewRepresentable {
#Binding var status: Bool
func makeUIView(context: Context) -> UIActivityIndicatorView {
let activityIndicator = UIActivityIndicatorView()
activityIndicator.style = .large
return activityIndicator
}
func updateUIView(_ uiView: UIViewType, context: Context) {
print("Hello")
}
}
import SwiftUI
struct ChartView: View {
#State var status = false
var body: some View {
VStack{
Spacer()
Button(action: {
status = !status
}) {
Text("Change")
}
BindingTest(status: $status)
}
}
}
I am using Xcode 12.5 and Swift 5.4
Looks like SwiftUI is doing some cleverness under the hood that isn't immediately obvious to us. Because you don't actually use your binding in updateUIView, it's not actually getting called.
However, if you update your code to the following:
func updateUIView(_ uiView: UIViewType, context: Context) {
print("Hello \(status)")
}
then you'll see that it does, in fact, get called.
PS - you can use status.toggle() instead of status = !status
Related
I have a view Structure as follows:
ZStack {
UIKitView //PKCanvasView
.zIndex(1)
SwiftUIView //SwiftUI view which has a gesture recogniser
.zIndex(0)
}
I wanted to know what the best way was to allow simultaneous interaction to the SwiftUI view below the UIKit view.
This would work best if the SwiftUI view had a higher priority gesture over the UIKit view on top.
My UIKit representable view is as follows:
struct CanvasView: UIViewRepresentable {
#Binding var canvasView: PKCanvasView
#Binding var isFirstResponder: Bool
#State var toolPicker = PKToolPicker()
func makeUIView(context: Context) -> PKCanvasView {
canvasView.drawingPolicy = .pencilOnly
canvasView.backgroundColor = .clear
canvasView.isOpaque = false
canvasView.alwaysBounceVertical = true
canvasView.alwaysBounceHorizontal = true
toolPicker.setVisible(true, forFirstResponder: canvasView)
toolPicker.addObserver(canvasView)
//canvasView.becomeFirstResponder()
return canvasView
}
func updateUIView(_ uiView: PKCanvasView, context: Context) {
if isFirstResponder != uiView.isFirstResponder {
if isFirstResponder {
uiView.becomeFirstResponder()
} else {
uiView.resignFirstResponder()
}
}
}
}
Ideally I would like to handle the layering of views in SwiftUI as in reality a have multiple SwiftUI views below the canvas view which when held & dragged (that is the compound gesture I am applying to them) should appear above it. (This is achieved using dynamic zIndexs). I'm not comfortable enough with UIKit to know how to do this there with subviews and a UIViewRepresentable structure.
Any help would be greatly appreciated!
I am working on a SwiftUI project, the functionalities it required is to make a Rich Text Editor on IOS.
The approach I am following is fairly simple, I used cbess/RichTextEditor link originally written in UIKit and import it into SwiftUI. To run the imported UIView, I wrap the view inside one UIViewRpresentable and add it into the ContentView struct of SwiftUI.
Now, I want to publish the data inside UIView and assign it to one of #state ContentView owns.
The code structure look similar to this:
For the ContentView (SwiftUI)
struct ContentView: View {
#State var textHtml: String = "" //I want all changes come from UIView be stored inside this
var body: some View {
VStack {
Cbess(
frameEditor: CGRect(x: 0, y: 40, width: 360, height: 400)
)
}
}
}
For the UiViewRepresentable
struct Cbess : UIViewRepresentable{
let frameEditor : CGRect
func makeUIView(context: Context) -> UIView {
let frameEditor = RichEditorView(frame: frameEditor)
let uiView : UIView = UIView()
uiView.addSubview(editorView)
return uiView
}
func updateUIView(_ uiView: UIView, context: Context) {
}
}
For the UiView(Simplified)
#objcMembers open class RichEditorView: UIView, {
var contentHTML : String // This variable get updated regularly
}
One additional question is that I want to make a Rich Text Editor by solely SwiftUI. How can I achieve it? Can you give me some keywords? Some Repo?
Any help is very appreciated! Thanks for read this whole question.
Use #Binding and delegate.
UIViewRepresentable view
struct Cbess : UIViewRepresentable {
#Binding var textHtml: String
func makeCoordinator() -> Coordinator {
Coordinator(self)
}
func makeUIView(context: Context) -> RichEditorView {
let editorView = RichEditorView()
editorView.delegate = context.coordinator
return editorView
}
func updateUIView(_ uiView: RichEditorView, context: Context) {
}
class Coordinator: NSObject, RichEditorDelegate {
var parent: Cbess
init(_ parent: Cbess) {
self.parent = parent
}
// Use delegate here
func richEditor(_ editor: RichEditorView, contentDidChange content: String) {
self.parent.textHtml = content
print(content)
}
}
}
Your content view:
struct ContentView: View {
#State var textHtml: String = ""
var body: some View {
VStack {
Cbess(textHtml: $textHtml)
.frame(width: 360, height: 400)
Text("Print----\n\(textHtml)")
}
}
}
I'm using SwiftUI and RealityKit to make an AR app. I am trying to transition from a nonAR camera mode ARView to a regular ARView using a NavigationLink, but I'm running into a SIGABRT and see the following error whenever I select the link:
validateTextureDimensions, line 1227: error 'MTLTextureDescriptor has width (4294967295) greater than the maximum allowed size of 16384.'
I've reproduced this behavior in a fresh RealityKit app with these simple views:
// ContentView (nonAR ARView)
import SwiftUI
import RealityKit
struct ContentView : View {
var body: some View {
NavigationView {
VStack {
ARViewContainer().edgesIgnoringSafeArea(.all)
NavigationLink(destination: ContentView2()) {
Text("Go!")
}
}
}
}
}
struct ARViewContainer: UIViewRepresentable {
func makeUIView(context: Context) -> ARView {
let arView = ARView(frame: .zero, cameraMode: .nonAR, automaticallyConfigureSession: true)
return arView
}
func updateUIView(_ uiView: ARView, context: Context) {}
}
// ContentView2 (AR ARView)
import SwiftUI
import RealityKit
struct ContentView2: View {
var body: some View {
return ARViewContainer2().edgesIgnoringSafeArea(.all)
}
}
struct ARViewContainer2: UIViewRepresentable {
func makeUIView(context: Context) -> ARView {
let arView = ARView(frame: .zero)
return arView
}
func updateUIView(_ uiView: ARView, context: Context) {}
}
It seems like some cleanup needs to be performed with the first nonAR ARView before navigating but I'm not sure what the best way to manage that is. I saw the answer in this post and I tried adding a #Binding to ARViewContainer that I set in the parent view to flag to that the uiview should be removed in updateUIView before navigating, but I'm still hitting the crash :/ Any help here would be greatly appreciated!
I ended up resolving this for the time being by using a SceneKit view in place of the nonAR AR view.
I try to create a Facebook button but when I call it in the VStack it says: 'Referencing initializer 'init(alignment:spacing:content:)' on 'VStack' requires that 'FBLoginButton' conform to 'View'', So why isn't it working because FBLoginButton is a button after all.
here's the code:
struct ContentView: View {
var body: some View {
VStack(alignment: .center) {
FBLoginButton(frame: CGRectMake(0, 0, 100, 100), permissions: ["email"])
}
}
}
First of all needs to create class which implements UIViewRepresentable protocol.
import FBSDKLoginKit
import SwiftUI
struct FacebookLoginButton: UIViewRepresentable {
typealias UIViewType = FBLoginButton
func makeUIView(context: Context) -> UIViewType {
FBLoginButton()
}
func updateUIView(_ uiView: FBLoginButton, context: Context) { }
}
After all this FacebookLoginButton can be used in SwiftUI like this:
var body: some View {
FacebookLoginButton()
}
Facebook SDK still uses UIKit, and FBLoginButton is UIView from UIKit, not View" from SwiftUI.
You need to use loginmanger which does most of the work and embed it inside swiftui Button.
Check link below which will be helpful.
https://gist.github.com/zydeico/8bb9c772e944d1490335e1615aaf4960
I'm trying to find a way to trigger an action that will call a function in my UIView when a button gets tapped inside swiftUI.
Here's my setup:
foo()(UIView) needs to run when Button(SwiftUI) gets tapped
My custom UIView class making use of AVFoundation frameworks
class SomeView: UIView {
func foo() {}
}
To use my UIView inside swiftUI I have to wrap it in UIViewRepresentable
struct SomeViewRepresentable: UIViewRepresentable {
func makeUIView(context: Context) -> CaptureView {
SomeView()
}
func updateUIView(_ uiView: CaptureView, context: Context) {
}
}
SwiftUI View that hosts my UIView()
struct ContentView : View {
var body: some View {
VStack(alignment: .center, spacing: 24) {
SomeViewRepresentable()
.background(Color.gray)
HStack {
Button(action: {
print("SwiftUI: Button tapped")
// Call func in SomeView()
}) {
Text("Tap Here")
}
}
}
}
}
You can store an instance of your custom UIView in your representable struct (SomeViewRepresentable here) and call its methods on tap actions:
struct SomeViewRepresentable: UIViewRepresentable {
let someView = SomeView() // add this instance
func makeUIView(context: Context) -> SomeView { // changed your CaptureView to SomeView to make it compile
someView
}
func updateUIView(_ uiView: SomeView, context: Context) {
}
func callFoo() {
someView.foo()
}
}
And your view body will look like this:
let someView = SomeViewRepresentable()
var body: some View {
VStack(alignment: .center, spacing: 24) {
someView
.background(Color.gray)
HStack {
Button(action: {
print("SwiftUI: Button tapped")
// Call func in SomeView()
self.someView.callFoo()
}) {
Text("Tap Here")
}
}
}
}
To test it I added a print to the foo() method:
class SomeView: UIView {
func foo() {
print("foo called!")
}
}
Now tapping on your button will trigger foo() and the print statement will be shown.
M Reza's solution works for simple situations, however if your parent SwiftUI view has state changes, every time when it refreshes, it will cause your UIViewRepresentable to create new instance of UIView because of this: let someView = SomeView() // add this instance. Therefore someView.foo() is calling the action on the previous instance of SomeView you created, which is already outdated upon refreshing, so you might not see any updates of your UIViewRepresentable appear on your parent view.
See: https://medium.com/zendesk-engineering/swiftui-uiview-a-simple-mistake-b794bd8c5678
A better practice would be to avoid creating and referencing that instance of UIView when calling its function.
My adaption to M Reza's solution would be calling the function indirectly through parent view's state change, which triggers updateUIView :
var body: some View {
#State var buttonPressed: Bool = false
VStack(alignment: .center, spacing: 24) {
//pass in the #State variable which triggers actions in updateUIVIew
SomeViewRepresentable(buttonPressed: $buttonPressed)
.background(Color.gray)
HStack {
Button(action: {
buttonPressed = true
}) {
Text("Tap Here")
}
}
}
}
struct SomeViewRepresentable: UIViewRepresentable {
#Binding var buttonPressed: Bool
func makeUIView(context: Context) -> SomeView {
return SomeView()
}
//called every time buttonPressed is updated
func updateUIView(_ uiView: SomeView, context: Context) {
if buttonPressed {
//called on that instance of SomeView that you see in the parent view
uiView.foo()
buttonPressed = false
}
}
}
Here's another way to do it using a bridging class.
//SwiftUI
struct SomeView: View{
var bridge: BridgeStuff?
var body: some View{
Button("Click Me"){
bridge?.yo()
}
}
}
//UIKit or AppKit (use NS instead of UI)
class BridgeStuff{
var yo:() -> Void = {}
}
class YourViewController: UIViewController{
override func viewDidLoad(){
let bridge = BridgeStuff()
let view = UIHostingController(rootView: SomeView(bridge: bridge))
bridge.yo = { [weak self] in
print("Yo")
self?.howdy()
}
}
func howdy(){
print("Howdy")
}
}
Here is yet another solution! Communicate between the superview and the UIViewRepresentable using a closure:
struct ContentView: View {
/// This closure will be initialized in our subview
#State var closure: (() -> Void)?
var body: some View {
SomeViewRepresentable(closure: $closure)
Button("Tap here!") {
closure?()
}
}
}
Then initialize the closure in the UIViewRepresentable:
struct SomeViewRepresentable: UIViewRepresentable {
// This is the same closure that our superview will call
#Binding var closure: (() -> Void)?
func makeUIView(context: Context) -> UIView {
let uiView = UIView()
// Since `closure` is part of our state, we can only set it on the main thread
DispatchQueue.main.async {
closure = {
// Perform some action on our UIView
}
}
return uiView
}
}
#ada10086 has a great answer. Just thought I'd provide an alternative solution that would be more convenient if you want to send many different actions to your UIView.
The key is to use PassthroughSubject from Combine to send messages from the superview to the UIViewRepresentable.
struct ContentView: View {
/// This will act as a messenger to our subview
private var messenger = PassthroughSubject<String, Never>()
var body: some View {
SomeViewRepresentable(messenger: messenger) // Pass the messenger to our subview
Button("Tap here!") {
// Send a message
messenger.send("button-tapped")
}
}
}
Then we monitor the PassthroughSubject in our subview:
struct SomeViewRepresentable: UIViewRepresentable {
let messenger = PassthroughSubject<String, Never>()
#State private var subscriptions: Set<AnyCancellable> = []
func makeUIView(context: Context) -> UIView {
let uiView = UIView()
// This must be run on the main thread
DispatchQueue.main.async {
// Subscribe to messages
messenger.sink { message in
switch message {
// Call funcs in `uiView` depending on which message we received
}
}
.store(in: &subscriptions)
}
return uiView
}
}
This approach is nice because you can send any string to the subview, so you can design a whole messaging scheme.
My solution is to create an intermediary SomeViewModel object. The object stores an optional closure, which is assigned an action when SomeView is created.
struct ContentView: View {
// parent view holds the state object
#StateObject var someViewModel = SomeViewModel()
var body: some View {
VStack(alignment: .center, spacing: 24) {
SomeViewRepresentable(model: someViewModel)
.background(Color.gray)
HStack {
Button {
someViewModel.foo?()
} label: {
Text("Tap Here")
}
}
}
}
}
struct SomeViewRepresentable: UIViewRepresentable {
#ObservedObject var model: SomeViewModel
func makeUIView(context: Context) -> SomeView {
let someView = SomeView()
// we don't want the model to hold on to a reference to 'someView', so we capture it with the 'weak' keyword
model.foo = { [weak someView] in
someView?.foo()
}
return someView
}
func updateUIView(_ uiView: SomeView, context: Context) {
}
}
class SomeViewModel: ObservableObject {
var foo: (() -> Void)? = nil
}
Three benefits doing it this way:
We avoid the original problem that #ada10086 identified with #m-reza's solution; creating the view only within the makeUIView function, as per the guidance from Apple Docs, which state that we "must implement this method and use it to create your view object."
We avoid the problem that #orschaef identified with #ada10086's alternative solution; we're not modifying state during a view update.
By using ObservableObject for the model, we can add #Published properties to the model and communicate state changes from the UIView object. For instance, if SomeView uses KVO for some of its properties, we can create an observer that will update some #Published properties, which will be propagated to any interested SwiftUI views.