We Keep Coding

How can I find out that JavaFX Image loading in background has failed?

Question in short:
How can I find out, that background loading of image has failed just before imageView.setImage(image) results in showing an empty picture although image.isError==false and image.getException==null?
Background:
In my simple JavaFX based photoviewer app I use a TableView() for showing a directory with jpg files. Whenever one selects an entry in the table the picture is loaded using javafx Image class and is shown using an ImageView.
I load the photos in background using true in the parameter of the Image-constructor.
Once a photo is loaded I keep it in a list ("Cache") for faster "showing again"
Here the code snippet:
public Object getMediaContent() {
Image image = (Image) content;
if (!isMediaContentValid()) { //if not already loaded or image in cache is invalid
try {
System.out.println("getMediaContent loading " + fileOnDisk);
content = new Image(fileOnDisk.toUri().toString(), true); //true=load in Background
} catch (Exception e) {
//will not occur with backgroundLoading: image.getException will get the exception
System.out.println("Exception while loading:");
e.printStackTrace();
}
} else {
System.out.println(fileOnDisk.toString() + "in Cache :-)...Error="+ image.isError() + " Exception=" + image.getException());
}
return content;
}
In isMediaContentValid() I test
if the image is already in the cache's list
and if image.isError() is false
and if image.getException() is null
The problem:
When the user selects photos very quickly (e.g. by using cursor down key) images are still loaded in background (for the cache) while the next photo's loading has already started.
My simple chache algorithm has problems to find out when I run out of memory because there might be enough memory when loading is started but not to complete all background tasks.
But I expected that this is not a problem because image.isError() would report true or image.getException() would be != null in this case. So I could free memory before retrying.
But isError() reports false, getException() reports null and the image is shown "empty" in imageView :-(
The question:
How can I find out, that background loading of image has failed just before imageView.setImage(image)?

How can I find out, that background loading of image has failed just before imageView.setImage(image)?
This is not possible. The whole point of loading the image in the background is that it's done asynchronously. There is no guarantee that the exception already occured at the time the method returns. You need to use a listener to the error property to be notified of a failure to load the image.
Example
Image image = new Image("https://stackoverflow.com/abc.jpg", true); // this image does not (currently) exist
image.errorProperty().addListener(o -> {
System.err.println("Error Loading Image " + image.getUrl());
image.getException().printStackTrace();
});

define a Custom Property as a Data Object

Found the issue on my payment Input control. There was a small computed Text field that was failing but not throwing an error. Just stopped the whole process. In any case removed the computedText and it now works. The compositeData formula that returns pItem to the custom control still fires way to often but I can't figure out how to stop that. It is all memory resident so it is probably not a major performance hit, but still.....
This question is a follow up to my previous question and I will try to refine the issue
defining an object property in a compositeData on a custom control
Here is a picture of what I am trying to do:
The repeat control is bound to an arrayList generated by the Java method Payments.getAllItems(LinkKey) and that works correctly. The button in the repeat is fairly simple it just setts the viewScope.vsShowPayment = true and vsRIndex to the repeat Index value so I know which element in the ArrayList we are working with. It then does a refresh of the panelPaymentContainer which hides the repeat and renders the custom control ccTestPayment.
ccTestPayment has a custom property called pItem of the type java.lang.Object with this code:
<xc:ccTestPaymentInput rendered="#{javascript:(viewScope.vsShowPayment)}">
<xc:this.pItem><![CDATA[#{javascript:try{
var debug:Boolean = true;
if (debug) print("Open existing row = " + viewScope.vsRIndex)
rIndex = parseInt(viewScope.vsRIndex.toString());
if (debug) print("rIndex = " + rIndex);
pItem = Payments.getItem(rIndex);
return pItem;
}catch(e){
print("Failure in Custom Prop of add item " + e.toString());
return null;
}}]]></xc:this.pItem>
</xc:ccTestPaymentInput>
the method in the class Payments Payments.getItem(rIndex) then returns the PaymentItem Object from the ArrayList of PaymentItems and displays them in custom control. the fields in the custom control are bound to compositeData.pItem.getPaymentDate etc and to this point everything is cool.
I can edit any of the fields on the custom control and that all works fine. However, when I press the "Save" button none of the code in it gets executed.
try{
print("Start Payment save");
var debug:Boolean = true;
var pos:Integer = parseInt(viewScope.vsRIndex.toString());
if (debug) print("Working with pos = " + pos + " Call saveThisItem");
if (Payments.saveThisItem(compositeData.pItem , pos)){
if (debug) print("save Payments Worked ");
}else{
if (debug) print("save Payments FAILED ");
}
}catch(e){
print("payment save Error " + e.tostring);
}finally{
viewScope.vsExpPayDate = null;
viewScope.vsShowPayment = false;
viewScope.remove("vsRIndex");
viewScope.remove("vsGotItem")
}
None of the print statements get fired. I suspect it has something to do how pItem gets defined. the code behind the custom property gets fired over and over again and I'm wondering if that is getting in the way.

The reason that the save was not working was that there was a computed Text field on the control that generated an error. The problem was that there was no error message reported on the client nor the console. After a lot of head scratching I noticed the text filed was no longer displaying the value it was supposed to. Deleted the field and the save and everything else started to work.
On the issue of the number of times the processes are called I think I have resolved many of them. I'm moving the control ccTestPaymentInput.xsp inside the repeat. It will now have direct access to the 'current' PaymentItem Object so I can access teh repeat var=pItem which is teh PaymentItem object I want to work with. Clean and far simpler than what I was doing. The only refreshes necessary are the ones related to the rpeat control and there is not much that I can do about that.

Capturing Android screenshot with appium

I currently have something working to capture screenshots on iOS for appium tests using java and junit. Before a test finishes it runs this code to get the last visible thing on the screen before killing the connection
#After
public void tearDown() throws Exception {
if (platform.equals(iOS)) {
captureScreenshot(testName.getMethodName());
}
driver.quit();
}
#SuppressWarnings("Augmenter")
public void captureScreenshot(String testName) {
String imagesLocation = "target/surefire-reports/screenshot/" + platform + "/";
new File(imagesLocation).mkdirs(); // Insure directory is there
String filename = imagesLocation + testName + ".jpg";
try {
Thread.sleep(500);
WebDriver augmentedDriver = new Augmenter().augment(driver);
File scrFile = ((TakesScreenshot)augmentedDriver).getScreenshotAs(OutputType.FILE);
FileUtils.copyFile(scrFile, new File(filename), true);
} catch (Exception e) {
System.out.println("Error capturing screen shot of " + testName + " test failure.");
// remove old pic to prevent wrong assumptions
File f = new File(filename);
f.delete(); // don't really care if this doesn't succeed, but would like it to.
}
}
This sorta works for android but lately it has completely killed the running test but this might have to do with the device it is trying to get the snapshot of. Before it would save the file but the image would come up blank or broken.
Anyone know how to get image/screen capturing working on android using appium? Perhaps something with UIAutomator?

You can use this method:
public void screenshot(String path_screenshot) throws IOException{
File srcFile=driver.getScreenshotAs(OutputType.FILE);
String filename=UUID.randomUUID().toString();
File targetFile=new File(path_screenshot + filename +".jpg");
FileUtils.copyFile(srcFile,targetFile);
}
It works fine for me.

I happened to find this via Google search with nearly the same problem - Appium screenshots in the Android emulator come up blank. I'm using both the 'native' method much like you describe above - and - the method within the ATU framework.
WebElement appArea = wd.findElementByXPath("//android.widget.LinearLayout[1]/android.widget.FrameLayout[1]/android.widget.LinearLayout[1]");
ATUReports.add("Main Screen Area Screenshot", LogAs.INFO, new CaptureScreen(appArea));
and both come back blank/transparent. The marginally good news is that the dimension of that blank/transparent image is exactly the size I want to capture - avoiding the status bar where date diffs would cause error on image comparison. That said, it's not much use without actual viewable pixels (for eventual matching of object area grabs to validated baseline images).
I tried setting context to "NATIVE_APP" and nothing seems to help. I am - in a word - vexed. If you've made any progress with this I'd love to read/hear how you got it working. Maybe the next step is to go to the Appium Google Group.
EDIT: I found the core issue in my case - using HAXM acceleration causes the blank screen shots. Note this also affects test run on physical devices if the base device profile set in the test's capabilities is defined with "Host GPU" selected.

How to load a Google maps static map using Picasso?

In my Android app I use Picasso to load images. This normally works perfectly well.
Today I tried loading a static image from the google maps api, but this doesn't seem to work. When I open the example link as provided on their info page, I get to see the static map image perfectly well. When I load it in my Android app using the line below, I get nothing at all.
Picasso.with(getContext()).load("http://maps.googleapis.com/maps/api/staticmap?center=Brooklyn+Bridge,New+York,NY&zoom=13&size=370x250&maptype=roadmap%20&markers=color:blue|label:S|40.702147,-74.015794&markers=color:green|label:G|40.711614,-74.012318%20&markers=color:red|color:red|label:C|40.718217,-73.998284&sensor=false").into(mapView);
I also tried to download the image and uploading it to my personal webspace, from which it loads perfectly well, but somehow, it doesn't seem to load directly from the direct google API url.
Does anybody know why this is so, and how I can solve it?

The only programmatic point-of-failure that comes to mind is in parsing the URI. Looking at the current Picasso code (https://github.com/square/picasso/blob/master/picasso/src/main/java/com/squareup/picasso/Picasso.java) I see the following:
public RequestCreator load(String path) {
if (path == null) {
return new RequestCreator(this, null, 0);
}
if (path.trim().length() == 0) {
throw new IllegalArgumentException("Path must not be empty.");
}
return load(Uri.parse(path));
}
So I'd first debug
Uri.parse("http://maps.googleapis.com/maps/api/staticmap?center=Brooklyn+Bridge,New+York,NY&zoom=13&size=370x250&maptype=roadmap%20&markers=color:blue|label:S|40.702147,-74.015794&markers=color:green|label:G|40.711614,-74.012318%20&markers=color:red|color:red|label:C|40.718217,-73.998284&sensor=false")
and see what that Object looks like. Does it drop or confuse any of your parameters?
If that doesn't lead you anwhere, try downloading the file manually using a HttpClient [or similar]. Then at least you can fully debug the request/response.
Also, I know Google maps has some limits -- are you sure you haven't reached them?

replace http with https
replace | with %7C
add api key

The .loadMap() function has many declared variables. This is the heart of the whole process.
So what is required for the static maps API to give us an image is that we make an http request with a given url, for which an image response (URL) is received. Let us run through the meaning and utility of these variables. Yes, all of them have a completely different meaning!
The mapUrlInitial variable is always the same while making an API call. It has a query of center ( ?center ) which specifies that we want the location to be centered in the map.
The mapUrlProperties variable contains a string where you control the actual zooming of the image response you will get, the size ofthe image and the color of the marker which will point out our place.
The mapUrlMapType variable is a string where you can actually determine the marker size you want and the type of the map. We are using a roadtype map in the app.
Finally latLong is a string which concatenates the latitude and the longitude of the place we want to pinpoint!
We then concatenate all of these strings to form a feasible Url. The Url is then loaded as we have seen above, in the Picasso code. One thing we can notice is that an event object is always required for all of this to happen, because we are able to fetch the position details using the event object! Final Code:-
fun loadMap(event: Event): String{
//location handling
val mapUrlInitial = “https://maps.googleapis.com/maps/api/staticmap?center=”
val mapUrlProperties = “&zoom=12&size=1200×390&markers=color:red%7C”
val mapUrlMapType = “&markers=size:mid&maptype=roadmap”
val latLong: String = “” +event.latitude + “,” + event.longitude
return mapUrlInitial + latLong + mapUrlProperties + latLong + mapUrlMapType
}
//load image
Picasso.get()
.load(loadMap(event))
.placeholder(R.drawable.ic_map_black_24dp)
.into(rootView.image_map)

Android Libgdx TileAtlas

I am trying to add a Map to my libgdx app as a proof of concept. It seems that no matter how I make a packfile, the com.badlogic.gdx.graphics.g2d.tiled.TileAtlas constructor TileAtlas(TiledMap map, FileHandle inputDir) will not correctly read it. My Tile Map is simple and has only 2 tiles, and both the external gui and internal system will generate a packed file.
Here's the issue, either I name the packfile with a filename to match one of my images to satisfy line 2 below, or the method errors out. If I add 2 packfiles, one for each name of an image in my tile set, I find the Atlas isn't constructed correctly in memory. What am I missing here? Should there only ever be one tile in a tilemap?
Code from Libgdx:
for (TileSet set : map.tileSets) {
FileHandle packfile = getRelativeFileHandle(inputDir, removeExtension(set.imageName) + " packfile");
TextureAtlas textureAtlas = new TextureAtlas(packfile, packfile.parent(), false);
Array<AtlasRegion> atlasRegions = textureAtlas.findRegions(removeExtension(removePath(set.imageName)));
for (AtlasRegion reg : atlasRegions) {
regionsMap.put(reg.index + set.firstgid, reg);
if (!textures.contains(reg.getTexture())) {
textures.add(reg.getTexture());
}
}
}

com.badlogic.gdx.graphics.g2d.tiled --> It looks like you're using the old tiled API. I don't even think that package exists anymore, so you should probably download a newer version.
Check out this blog article. I haven't used the new API yet, but at a quick glance it looks much easier to load maps.