Hope everyone knows the basic Libgdx game in this link:
https://github.com/libgdx/libgdx/wiki/A-simple-game
In the simple game,rain drops are falling randomly from the top,that is of one type.
In my case,I want to implement the same thing,as drops.The difference is that,there are four type of drops.When it collides with another object,four should show four different characteristics.
In short,I created a Drop class with its object and array and can spawn drop objects too.
Drop drop=new Drop();
private Array<Drop> drops=new Array<Drop>();
But what if I want to define individual attributes for drops in this case?
I want the drop object to be categorized of four types. For example have different colors yellow,red,green and blue.Also four kind of drops should fall randomly from top.
In which way,and how I should implement such a concept?
It would be very helpful if I get some idea on this.
Edit:
I have an abstract class Drop and .I am not passing texture and rectangle in the constructor.Because graphics are not available for the project right now.I am planning to assign it in a later stage.As of now,everything I am doing it with shaperenderer.
I have create() method for three distinct colored drops in my objectFactory class where I used to create all objects:
public YellowDrop createYellow(){
YellowDrop yellow = new YellowDrop();
yellow.setSize(100f,100f);
yellow.setPosition(MathUtils.random(0, 800),1280);
return yellow;
}
public RedDrop createRed(){
RedDrop red = new RedDrop();
red.setSize(100f,100f);
red.setPosition(MathUtils.random(0, 800),1280);
return red;
}
public GreenDrop createGreen(){
GreenDrop green = new GreenDrop();
green.setSize(100f,100f);
green.setPosition(MathUtils.random(0, 800),1280);
return green;
}
And I written createRandomDrop() code like this(For me,your code is confusing though.How can I use that colon symbol there in a method?I never used it.)
private Block createRandomDrop() {
switch (MathUtils.random(0, 3)) {
case 0 :
System.out.println("000000");
return objectFactory.createYellow();
case 1 :
return objectFactory.createGreen();
case 2 : return objectFactory.createRed();
default:
return objectFactory.createGreen();
}
}
Here I am confused about how to return the individual drop object.
Even though I have written the call like this:
if(TimeUtils.nanoTime() - lastDropTime > 1000000000) {
drops.add(createRandomDrop());
}
private void updateBlocks(float delta) {
Iterator<Drop> iter = drops.iterator();
while(iter.hasNext()) {
Drop b = iter.next();
b.update(delta);//moving drop from top to bottom in update()
if(b.getY()<0) iter.remove();
}
// System.out.println(drops.size);
}
At last I am drawing the shaperenderer for all array elements:
for (Drop b : drops) {
b.drawDebug(shapeRenderer);}
This drawdebug() is in Drop class and I am overriding it
public void drawDebug(ShapeRenderer shapeRenderer) {
shapeRenderer.setColor(Color.YELLOW);
shapeRenderer.rect(collisionRectangle.x, collisionRectangle.y,
collisionRectangle.width,
collisionRectangle.height);
}
problem is that,too many drops are getting created and overlapping.May be the switch case creates the problem.
What is your problem? This is simple.
Create 4 children of Drop class. yellowDrop, greenDrop, etc... And create collide() method in Drop class which will implement same functionality for all children.
Something like this:
abstract class Drop {
abstract Texture tex;
abstract Rectangle rect;
Drop(Texture tex, Rectangle rect){
// initialize values here
}
void showCollideAnimation(){ // example. here you can create your own void to show different characteristics.
tex.blabla()
}
void collide(){
showCollideAnimation();
}
// also draw, update methods etc... Hope you got that they should do functionality same for all children
}
And then create child that will override showCollideAnimation method.
class GreenDrop extends Drop {
#override
void showCollideAnimation(){
.... // here you pass effect for Green child
}
}
To chose random child you can just create function that returns Drop
Drop createRandomDrop(Rectangle rect): Drop {
switch (MathUtils.random(0, 3)) {
case 0 : return new YellowDrop(Texture("pathToYellowTexture"), rect) break;
case 1 : return new GreenDrop(Texture("pathToGreenTexture"), rect) break;
// etc.
}
}
Now you can fill ArrayList of Drops.
private ArrayList<Drop> drops = new ArrayList<Drop>();
in render method:
if(TimeUtils.nanoTime() - lastDropTime > 1000000000) {
drops.add(createRandomDrop(Texture(""), Rectangle())); // specify texture and rectangle yourself
}
Unity 5.4
Building a Unity WebGL application (not a game) that handles all 3D content on the Unity side of things, and all the UI is built using HTML/CSS/JS. By default, WebGLInput.captureAllKeyboardInput is set to true, which causes any input fields that require keyboard (text) input to be broken, as any keyboard controls are automatically being eaten up by Unity rather than going to the input field. Doing
#if !UNITY_EDITOR && UNITY_WEBGL
WebGLInput.captureAllKeyboardInput = false;
#endif
fixes the issue with input fields, but causes unity to ignore ALL keyboard inputs, even after the element is focused, however adding
tabindex="1"
to the fixes it so that keyboard input on HTML input fields works when focused, and keyboard controls inside the Unity WebGL app works when focused as well. (this is all per the docs: https://docs.unity3d.com/ScriptReference/WebGLInput-captureAllKeyboardInput.html). So that's all working fine.
However, the Unity WebGL application is still causing issues with some input fields using MOUSE (not keyboard) controls. Namely I've noticed it on input type="range" fields (HTML5 sliders) and input type="number (using keyboard to type in numbers works, but mouse clicking up and down do not).
Is there any sort of workaround to fix this? I essentially need to prevent the Unity WebGL canvas to not automatically take all mouse inputs, unless the element is clicked/focused first (just like the way the keyboard controls work). Anything to change on the Unity side to fix this? Or do I need to write some custom JavaScript to handle all input and determine whether it's intended for the Unity scene or the HTML UI?
I was having the same issue but I believe I've found a workaround! Originally I was just going to use OnApplicationFocus to toggle WebGLInput.captureAllKeyboardInput. But OnApplicationFocus doesn't work with WebGL builds of Unity so I'm doing this instead.
I have a script named "GameControl" on a GameObject also named "GameControl" in the first scene when the game is loaded.
// In the Start function of this script I call a function on the webpage to let it know that the game has loaded.
void Start () {
#if (UNITY_WEBPLAYER || UNITY_WEBGL) && !UNITY_EDITOR
try {
Application.ExternalCall("GameControlReady");
} catch (System.Exception e) {
Debug.LogError("GameControlReady function not on webpage"+e);
}
#endif
}
// This function will be called from the webpage
public void FocusCanvas (string p_focus) {
#if !UNITY_EDITOR && UNITY_WEBGL
if (p_focus == "0") {
WebGLInput.captureAllKeyboardInput = false;
} else {
WebGLInput.captureAllKeyboardInput = true;
}
#endif
}
On the webpage, I have the following javascript:
var gameReady = false;
// Called by Unity in GameControl's start function
function GameControlReady () {
gameReady = true;
}
function FocusCanvas(focus) {
if (gameReady) {
SendMessage("GameControl", "FocusCanvas", focus);
}
}
And in the head section of the webpage I have the following
<script type='text/javascript'>
document.addEventListener('click', function(e) {
if (e.target.id == "canvas") {
// Clicked on canvas
FocusCanvas("1");
} else {
// Clicked outside of canvas
FocusCanvas("0");
}
});
</script>
Here I have AS3 to for uploading recorded sound file to server. When I test it in Flash it works properly (record sound and upload it and goes to next frame) , but in browser it doesn't work. It seems can't call myUpload but I don't why? Is it should be mouse event? Thanks.
function VOCWordToYourMp3()
{
setTimeout(startRecording,3000);
recorder.addEventListener(RecordingEvent.RECORDING, onRecording);
recorder.addEventListener(Event.COMPLETE, onRecordComplete);
}
function startRecording()
{
if (! recording)
{
recorder.record();
}
else if (recording)
{
recorder.stop();
recording = false;
}
}
function onRecording(e:RecordingEvent)
{
//
}
function onRecordComplete(e:Event):void
{
//
}
function renderWav(src, convertToMp3 = false)
{
//
function handleRenderTimer(e:TimerEvent)
{
//
}
function finishRender()
{
//
}
}
function makeIntoMp3(wav)
{
wav.position = 0;
mp3Encoder = new ShineMP3Encoder(wav);
mp3Encoder.addEventListener(Event.COMPLETE, mp3EncodeComplete);
mp3Encoder.addEventListener(ProgressEvent.PROGRESS, mp3EncodeProgress);
mp3Encoder.start();
function mp3EncodeProgress(e:ProgressEvent):void
{
//
}
function mp3EncodeComplete(e: Event):void
{
myUpload('sound1',mp3Encoder.mp3Data);
}
}
function myUpload(namefile:String,sba: ByteArray):void
{
//upload code
}
Update:
In Flash Player 10 and Actionscript 3.0, all the calls to URLLoader must be in the same callstack.
http://helpx.adobe.com/flash-player/kb/user-interaction-required-upload-download.html
What is same callstack mean?
I recommend using something like Vizzy: https://code.google.com/p/flash-tracer/ to debug your swf in the browser. You can put trace statements into the onRecordComplete() and myUpload() functions, etc, to see how far the code is getting, and to see if you are getting any new errors. You may have some kind of security sandbox error because you are running in the browser. Being able to see what that error is will help you figure out what to do next.
To use Vizzy, you need to have the debug player running in your browser, and to configure the right path to your log file. This is sometimes a little bit tricky, so I'll give you some tips:
Add a file to your home directory called mm.cfg, and populate it with these settings:
ErrorReportingEnable=1
AS3Verbose=0
TraceOutputFileEnable=1
AS3Trace=0
TraceOutputBuffered=0
AS3StaticProfile=0
Then in Vizzy you have to set up the path to the log. On my windows 7 machine it is here:
C:\Users\MyUserName\AppData\Roaming\Macromedia\Flash Player\Logs\flashlog.txt
It may be different depending on your operating system.
Good luck, and report back if you Vizzy gives you any new information.
I am using the JWPlayer source code for 6.0.2813 (http://developer.longtailvideo.com/trac/) and It seems the even though I have a movieclip and I added the jwplayer class as a child, the jwplayer creates itself as a child of the main stage, thus allowing it to expand to the bound of the stage and not my movieclip (which I want to be a resizeable/draggable container) in my flash.
I asked the forums for help but they said they never intended it this way and wasn't much help. I was hoping someone familar with the source code could point my in the right direction.
How can I get the JWPlayer to be contained to a movieclip?
Edit:
I made a little bit of progress.
I found the RootReference class in com/longtailvideo/jwplayer/utils/RootReference.as
public function RootReference(displayObj:DisplayObject) {
if (!RootReference.root) {
RootReference.root = displayObj.root;
RootReference.stage = displayObj.stage;
try {
Security.allowDomain("*");
} catch(e:Error) {
// This may not work in the AIR testing suite
}
}
}
And noticed that the RootReference.stage is where things get added as a child. RootReference.stage = displayObj.stage; where the player class object is sent as displayObj I changed it to be RootReference.stage = MovieClip(root).gui.video_container;
Then throughout the code RootReference.stage.stageHeight and RootReference.stage.stageWidth was used so I switched it to RootReference.stage.height and RootReference.stage.width. This got it to compile and now the video is within the container but the video's top left is center on my video_container's center and the video isn't resized to the size of my container, but rather the size of the video. Also the controls are completely messed up.
But I was able to resize and move the video around
Assuming my testing scenarios are representative of your use-cases, I think I managed to hack up a work around.
The gist of the approach is to replace RootReference.root and RootReference.stage with a fake stage object that you control. Because most of the jwPlayer classes refer to those static variables instead of their own root and stage variables, this seems to work, for the most part. What ended up being the most complicated issue was working with the Stage.stageVideo objects, which I think are the hardware accelerated video objects. These are always attached to the stage and thus weren't compatible with the fake stage object. The main issue with those is positioning, and I have it mostly worked out, but there is still one glitch which I'll describe later but it should be ok, now.
The jwPlayer embed script was causing a lot of problems, so to get started I switched to normal SWFObject-based embedding and added a javascript function to the page called getFlashvars() that returned the configuration settings. Then, I changed the com.longtailvideo.jwplayer.utils.Configger.loadExternal() method to the following:
private function loadExternal():void {
if (ExternalInterface.available) {
try {
//var flashvars:Object = ExternalInterface.call("jwplayer.embed.flash.getVars", ExternalInterface.objectID);
var flashvars:Object = ExternalInterface.call("getFlashvars");
if (flashvars !== null) {
// TODO: add ability to pass in JSON directly instead of going to/from a string
for (var param:String in flashvars) {
setConfigParam(param, flashvars[param]);
}
dispatchEvent(new Event(Event.COMPLETE));
return;
}
} catch (e:Error) {}
}
}
That's something you probably don't have to deal with per not using a webpage.
The fake stage class is called StageInterceptor and is a singleton. To apply it, there were minor changes in the RootReference class:
package com.longtailvideo.jwplayer.utils {
import flash.display.DisplayObject;
import flash.display.Stage;
import flash.system.Security;
// added --------
import somePackage.StageInterceptor;
/**
* Maintains a static reference to the stage and root of the application.
*
* #author Pablo Schklowsky
*/
/* Modified for a stackoverflow question: http://stackoverflow.com/questions/13325318/jwplayer-trying-to-bound-the-video-player-inside-my-own-container */
public class RootReference {
/** The root DisplayObject of the application. **/
public static var root:DisplayObject;
// altered --------
/** A reference to the stage. **/
private static var _stage:StageInterceptor;
// altered --------
public static function get stage():StageInterceptor {
return _stage;
}
public function RootReference(displayObj:DisplayObject) {
if (!RootReference.root) {
// altered --------
RootReference.root = StageInterceptor.singleton;
RootReference._stage = StageInterceptor.singleton;
try {
Security.allowDomain("*");
} catch(e:Error) {
// This may not work in the AIR testing suite
}
}
}
}
}
Also, I removed the set stage() setter method from the class.
In the document class, I have the following code. The MouseEvent.CLICK handler is to test positioning and re-sizing the movie. The only thing you really need are the first few lines:
// add StageInterceptor to the display tree
addChild(StageInterceptor.singleton);
// add the jwPlayer:
var p:Player = new Player();
StageInterceptor.singleton.addChild(p);
// for testing only:
stage.addEventListener(MouseEvent.CLICK, function(e:MouseEvent):void {
var stg:StageInterceptor = StageInterceptor.singleton;
if (e.altKey) {
// click + alt: ignored (so can play, etc)
return;
} else if (e.shiftKey) {
// click + shift: resizes
stg.width = e.stageX - stg.x;
stg.height = e.stageY - stg.y;
} else {
// click: moves video
stg.x = e.stageX;
stg.y = e.stageY;
}
});
I put StageInterceptor in the package somePackage. It looks like this:
package somePackage
{
import flash.display.Bitmap;
import flash.display.BitmapData;
import flash.display.InteractiveObject;
import flash.display.Shape;
import flash.display.Sprite;
import flash.events.Event;
import flash.geom.Point;
import flash.geom.Rectangle;
import flash.media.StageVideo;
public class StageInterceptor extends Sprite
{
private static var _singleton:StageInterceptor = new StageInterceptor();
public static function get singleton():StageInterceptor {
return _singleton;
}
private var _bg:Bitmap;
public function StageInterceptor()
{
super();
scrollRect = new Rectangle(0, 0, 500, 500);
var bmpData:BitmapData = new BitmapData(500, 500, false, 0);
_bg = new Bitmap(bmpData);
_bg.alpha = 0.1;
_bg.cacheAsBitmap = true;
addChild(_bg);
if (stage) {
initOnStage();
} else {
addEventListener(Event.ADDED_TO_STAGE, initOnStage);
}
}
private function initOnStage(e:Event = null):void {
if (e) {
removeEventListener(Event.ADDED_TO_STAGE, initOnStage);
}
stage.addEventListener(Event.RESIZE, onStageResized);
}
private function onStageResized(e:Event):void {
e.stopImmediatePropagation();
dispatchEvent(new Event(Event.RESIZE));
updateStageVids();
}
public function updateStageVids():void {
if (stage.stageVideos.length > 0) {
for each (var sv:StageVideo in stage.stageVideos) {
if (!sv.videoWidth || !sv.videoHeight) {
continue;
}
var rect:Rectangle = stretch(sv.videoWidth, sv.videoHeight, width, height);
rect.x = Math.max(0, x + 0.5 * (width - rect.width))
rect.y = Math.max(0, y + 0.5 * (height - rect.height));
sv.viewPort = rect;
}
}
}
override public function get width():Number {
return scrollRect.width;
}
override public function set width(value:Number):void {
if (value != width) {
_bg.width = value;
scrollRect = new Rectangle(0, 0, value, scrollRect.height);
dispatchEvent(new Event(Event.RESIZE));
updateStageVids();
}
}
override public function set height(value:Number):void {
if (value != height) {
_bg.height = value;
scrollRect = new Rectangle(0, 0, scrollRect.width, value);
dispatchEvent(new Event(Event.RESIZE));
updateStageVids();
}
}
override public function get height():Number {
return scrollRect.height;
}
public function get stageWidth():Number {
return scrollRect.width;
}
public function get stageHeight():Number {
return scrollRect.height;
}
public function get scaleMode():String {
return stage.scaleMode;
}
public function set scaleMode(value:String):void {
stage.scaleMode = value;
}
public function get displayState():String {
return stage.displayState;
}
public function set displayState(value:String):void {
stage.displayState = value;
}
public function get focus():InteractiveObject {
return stage.focus;
}
public function set focus(value:InteractiveObject):void {
stage.focus = value;
}
public function get stageVideos():* {
return stage.stageVideos;
}
override public function set x(value:Number):void {
if (value != x) {
super.x = value;
updateStageVids();
}
}
override public function set y(value:Number):void {
if (value != y) {
super.y = value;
updateStageVids();
}
}
/**
* Copied from com.longtailvideo.jwplayer.utils.Stretcher, modified to only
* do 'uniform' stretch and to return a Rectangle class.
**/
public static function stretch(elmW:Number, elmH:Number, availW:Number, availH:Number):Rectangle {
var scale:Number = Math.min(availW / elmW, availH / elmH);
elmW = Math.round(elmW * scale);
elmH = Math.round(elmH * scale);
return new Rectangle(0, 0, elmW, elmH);
}
}
}
The issue that remains has to do with the positioning of video instances when they are initialized. I think simply calling StageInterceptor.singleton.updateStageVids(); at the right point will do the trick, but I'm not sure. The edit below covers how this was addressed.
I'm not sure how well this will work if you're not using stageVideo. But, with any luck, this will move things in the right direction.
Edit:
I've updated the StageInterceptor class to do a better job scaling and positioning the video.
Also, it looks like the initial position of videos (at least when it's a stageVideo, is that what you're using?) can be corrected by a small edit in the com.longtailvideo.jwplayer.media.VideoMediaProvider class. Adding import somePackage.StageInterceptor; to the import statements at the top and then replacing this line (link to source):
_stage.viewPort = new Rectangle(_media.x,_media.y,_media.width,_media.height);
To:
StageInterceptor.singleton.updateStageVids();
So the method looks like:
/** Resize the video or stage.**/
override public function resize(width:Number, height:Number):void {
if(_media) {
Stretcher.stretch(_media, width, height, _config.stretching);
if (_stage) {
//_stage.viewPort = new Rectangle(_media.x,_media.y,_media.width,_media.height);
StageInterceptor.singleton.updateStageVids();
}
}
}
This should do the trick, but I haven't tested it for non stageVideos. And, this update also assumes you're playing videos progressively, not using RTMP media.
Edit:
To enable moving and resizing the player with non-StageVideo videos, but still progressively loaded, the contents of the com.longtailvideo.jwplayer.view.View.resizeMasker() method need to be either commented out or deleted:
protected function resizeMasker():void {
/*
if (_displayMasker == null)
setupDisplayMask();
_displayMasker.graphics.clear();
_displayMasker.graphics.beginFill(0, 1);
_displayMasker.graphics.drawRect(_components.display.x, _components.display.y, _player.config.width, _player.config.height);
_displayMasker.graphics.endFill();
*/
}
I also want to mention the open source version of the jwPlayer is governed under the Creative Commons license, as noted on their site:
JW Player 6 — Open Source Edition
The use of the JW Player Open Source edition is governed by a Creative Commons license. In short:
JW Player Open Source - You can use, modify, copy, and distribute this edition as long as it's for non-commercial use, you provide attribution, and share under a similar license.
The license summary and full text can be found here: CC BY-NC-SA 3.0
I'll address the jwPlayer portion of your question since flash is not my forte.
The problem here is jwPlayer is not a simple flash player, but also a HTML5 Multimedia Player as well.
Solution 1: SWFObject embedded in your Flash Object
Since jwPlayer is already compatible with SWFObject, use that (i.e., swfobject.js) as a mediator to load the player.swf file (aka jwPlayer) into your flash stage. To be sure, SWFObject acts as a container and will be the "bounded item" in your stage, as opposed to using jwPlayer directly.
Here is an online demo illustrating this solution except it's using a simple flash video player.
Flash Web Site with embedded SWFObject Logo Playing Music Video
Flash Web Site Documentation on swfObject
Note the HTML source page for that flash website shows RockOnFlashLogo.swf as the file being shown in the browsers entire viewport. Looking deeper, it's written in AS3.
Unlike jwPlayer v4 where many ideas floated on the internet to embed that version into flash websites due to its lax security, I think you'll have issues with current jwPlayer license checking, webpage Ready Event Listeners, and popular plugin integration... not to mention issues that may arise from streaming video content.
IMHO, the newer jwPlayer API and Player are intended to be use via webpage installation.
Solution 2: SWFObject On-Top of your Flash Object
This method treats jwPlayer how it was meant to be used; as a webpage installation. My motivation came from this online demo for Google Chrome Experiment | The Wilderness Downtown.
That demo places strategically synchronized browser windows on-top of the main browser window. Although the main browser window is in charge, all windows make up the entire experience. The same method, but flash flavored, can be done with your current project with excellent results.
Your Flash Object is in charge and contains within an interactive frame that is allocated for the jwPlayer webpage component. To be sure, this interactive frame for jwPlayer can accept relocation (e.g., via dragging frame edges) and resize (e.g, frame has resize-icon bottom-right) which is then relayed to the webpage component (i.e., player.swf) via SWFObject standard embed techniques (i.e., location and size set with jQuery).
Here's a basic cross-section of a webpage for three layered items:
The black layer is the HTML of the webpage. The red layer is your flash object which contains also has the built-in interactive frame shown in aqua. The green top layer is the SWFObject for jwPlayer's player.swf file.
Here's how it would work from jwPlayer to your flash object:
1. Webpage jwPlayer API has event listener active and can accept JavaScript.
2. Then jwPlayer API receives 'play' status from player, updates player event status.
3. Player Event Status is true for Play, and therefore triggers conditional if statement.
4. That if statement then transmits JavaScript to your flash object, indicating play mode is true.
5. Your flash object receives this JavaScript of play event, and dims the stage, less the aqua frame.
Here's how it would work from your flash object to jwPlayer:
1. The aqua frame has user interaction when it's moved to left side of stage.
2. Your AS3 code moves the aqua frame accordingly, while sending out JavaScript of that location.
3. Webpage receives JavaScript and invokes function to position jwPlayer player at new location.
Tip: If using a custom jwPlayer Skin, incorporate that skin theme into your flash object for a uniform look.
The benefit of this scenario is that jwPlayer maintains 100% integrity while both these flash objects work in tandem on your webpage. No hacks, no breakage, no unforeseen consequences, and no headaches... it's standard jwPlayer API and AS3 markup in use.
Hovering over the jwPlayer itself is 100% jwPlayer, while being bound to your flash object indirectly.
Per your written comments that JW Player will not have any access to JavaScript and will be using Firefox Source that is a specialized 3D Game/Chat Engine without access to any DOM elements outside the player, the best solution is to use JW Player Enterprise Edition.
That solution will put you in touch with the Marketing & Engineering department which can provide a turnkey solution to have JW Player integrated into your own product.
Click the image below which includes Licensing information as well:
I'm having issues with my Windows Store App launch. When I use the "close app gesture" (slide the app from top to bottom) and then launch the app again very fast, sometimes a blank black screen appears and when I click on it, it, the Start menu appears, and a "MoAppHang" event is logged.
My App_Launched event code is here:
protected override async void OnLaunched(LaunchActivatedEventArgs args)
{
if (args.PreviousExecutionState == ApplicationExecutionState.Terminated )
{
// Restore the saved session state only when appropriate
await SuspensionManager.RestoreAsync();
}
// Do not repeat app initialization when already running, just ensure that
// the window is active
if (args.PreviousExecutionState == ApplicationExecutionState.Running)
{
if (!string.IsNullOrEmpty(args.Arguments))
{
Frame f = Window.Current.Content as Frame;
if (f != null)
{
UseSecondaryTileNavigation(f, args.Arguments);
}
}
Window.Current.Activate();
return;
}
Frame rootFrame;
if (Window.Current.Content == null)
{
// Create a Frame to act as the navigation context and associate it with
// a SuspensionManager key
rootFrame = new Frame();
SuspensionManager.RegisterFrame(rootFrame, "AppFrame");
}
else
{
rootFrame = (Frame)Window.Current.Content;
}
if (!await DatabaseHelper.ExistsDatabase())
{
await DatabaseHelper.CreateDatabase();
}
if (rootFrame.Content == null)
{
// When the navigation stack isn't restored navigate to the first page,
// configuring the new page by passing required information as a navigation
// parameter
if (!rootFrame.Navigate(typeof(ItemsPage), "AllGroups"))
{
throw new Exception("Failed to create initial page");
}
}
if (!string.IsNullOrEmpty(args.Arguments))
{
UseSecondaryTileNavigation(rootFrame, args.Arguments);
}
// Place the frame in the current Window and ensure that it is active
if (Window.Current.Content == null)
{
Window.Current.Content = rootFrame;
}
Window.Current.Activate();
The UseSecondaryTileNavigation performs navigation when user opens the app using secondary tile (it basically uses the Frame parameter and Navigates it to the correct location using Frame.Navigate).
Where am I going wrong?
Thank you all!
It looks like you could potentially be doing a bit of time-consuming work in your launch handler. The first suggestion I'd make is to use a custom splash screen technique. That is, in your OnLaunched handler try to set the Window.Current.Content and activate the window and get out of that method ASAP. You could set Window.Current.Content to a page that just shows a loading progress bar - and handle your actual loading logic in there.
The next thing to look at - is what happens when your app is launched while it is still suspending? (Do you have a suspending handler?) Can your app handle the case when it is (re-)launched before a previous suspension / close completes?
I've noticed that it generally seems to take a few seconds before the app is fully closed (even though you close it using the drag down gesture).