I have IntentService that I use to upload data to server. When this upload goes wrong (sometimes when the user does not have reception or Internet connection) then I use exponential backoff to try and upload the information again and again. I do this inside the onHandleIntent method, which to my understanding runs in separate thread:
#Override
protected void onHandleIntent(Intent intent) {
Bundle extras = intent.getExtras();
if (extras != null) {
WebPostsRest posts = new WebPostsRest(HTTPVerb.POST, Uri.parse(CONSTS_WEB.URI_POST_NEW_POST), extras, this);
double backoffMillis = 0;
int backoffexp = 0;
int iTries = CONSTS_APP_GENERAL.NO_OF_POST_TRIES;
while (iTries > 0) {
mIsSuccess = posts.runRestNoReturnData();
if (mIsSuccess) {
iTries = 0;
} else {
if (backoffexp < 11) {
backoffMillis = Math.pow(2, backoffexp) * 1000;
backoffexp++;
}
try {
Thread.sleep((int) backoffMillis);
} catch (InterruptedException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
iTries--;
}
}
}
}
For some reason I get ANRs, which I assume come from this Thread.sleep part. Any idea why it happens and how to solve things? thank you.
I had the same problem and in my case it was causing by queue for the IntentService. Make sure you are not calling this intent service multiple times. It seems if first call of IntentService fail and use Thread.sleep() and at the same time more request are in waiting mode to use this IntentService then it throws ANR.
Related
In the clean code book is an example about using exceptions rather than return codes:
You either set an error flag or returned an error code.
public class DeviceController {
...
public void sendShutDown() {
DeviceHandle handle = getHandle(DEV1);
// Check the state of the device
if (handle != DeviceHandle.INVALID) {
// Save the device status to the record field
retrieveDeviceRecord(handle);
// If not suspended, shut down
if (record.getStatus() != DEVICE_SUSPENDED) {
pauseDevice(handle);
clearDeviceWorkQueue(handle);
closeDevice(handle);
} else {
logger.log("Device suspended. Unable to shut down");
}
} else {
logger.log("Invalid handle for: " + DEV1.toString());
}
}
...
}
Unfortunately, it's easy to forget. For this reason, it is better to throw an exception when you encounter an error. The calling code is cleaner. Its logic is not obscured by error handling.
public class DeviceController {
...
public void sendShutDown() {
try {
tryToShutDown();
} catch (DeviceShutDownError e) {
logger.log(e);
}
}
private void tryToShutDown() throws DeviceShutDownError {
DeviceHandle handle = getHandle(DEV1);
DeviceRecord record = retrieveDeviceRecord(handle);
pauseDevice(handle);
clearDeviceWorkQueue(handle);
closeDevice(handle);
}
private DeviceHandle getHandle(DeviceID id) {
...
throw new DeviceShutDownError("Invalid handle for: " + id.toString());
...
}
...
}
The code is better because two concerns that were tangled, the algorithm for device shutdown and error handling, are now separated. You can look at each of those concerns and understand them independently.
Now my question is not so much about the whether to use exceptions or return codes but I am wondering about the unused DeviceRecord in the second "good example", which uses exceptions.
Wouldn't it be enough to just call retrieveDeviceRecord(handle); to save the device status to the record field as in the "bad example code" but change the method to also throw an exception in case anything goes wrong during retrieveDeviceRecord(handle);? Or is there a purpose of returning DeviceRecord but not using it?
I am trying to parse some pois from a xml download from a server and I saw that it is done after the program continues in the main thread. I haven't found a way to solve it because I need it.
using System.Threading;
namespace XML_Parser
{
class XMLParserPOI_Wiki
{
private static XMLParserPOI_Wiki objSingle = new XMLParserPOI_Wiki();
public static XMLParserPOI_Wiki ObjSingle
{
get { return objSingle; }
set { objSingle = value; }
}
private List<POI> places;
public List<POI> Places
{
get { return places; }
}
private XMLParserPOI_Wiki()
{
}
public void parseWikitude(string url)
{
places = new List<POI>();
WebClient wc = new WebClient();
wc.DownloadStringCompleted += HttpsCompleted;
wc.DownloadStringAsync(new Uri(url));
}
private void HttpsCompleted(object sender, DownloadStringCompletedEventArgs e)
{
if (e.Error == null)
{
XDocument xdoc = XDocument.Parse(e.Result, LoadOptions.None);
XNamespace ns = "http://www.opengis.net/kml/2.2";
XNamespace ns2 = "http://www.openarml.org/wikitude/1.0";
var placemarkers = xdoc.Root.Descendants(ns + "Placemark");
places =
(from query in xdoc.Root.Descendants(ns + "Placemark")
select new POI
(
...
)).ToList();
System.Diagnostics.Debug.WriteLine("Lista");
System.Diagnostics.Debug.WriteLine(places.Count);
}
}
}
}
In my main class:
XMLParserPOI_Wiki parserXML = XMLParserPOI_Wiki.ObjSingle;
parserXML.parseWikitude("http://myurl.php");
System.Diagnostics.Debug.WriteLine("Lista de pois");
System.Diagnostics.Debug.WriteLine(parserXML.Places.Count);
for (int i = 0; i < parserXML.Places.Count; i++)
{
System.Diagnostics.Debug.WriteLine(parserXML.Places[i].getName());
}
It prints Lista de POis and 0, before Lista and X (number of pois)
I guess I should freeze main thread but I tried a couple of times with some examples and they didn't work.
Can you point me to any tutorial about this? More than get an answer I want to understand how to deal with this kind of operations
First of all, you don't want to block (freeze) the UI thread EVER!
This is called asynchronous programming. There are two things you can do to solve your problem (I recommend option 2!):
Use the classic callback model. You basically call some long operation on a background thread and give a function to it, to execute when the long operation is done. Here's how to do it in your case.
At the end of the HttpsCompleted method, invoke what you need on the UI Thread using:
Deployment.Current.Dispatcher.BeginInvoke(delegate() {
//The code here will be invoked on the UI thread
});
If you want to make the parseWikitude method reusable, you should pass an Action to it. This way you can call it from multiple places and tell it what to do on the UI thread when the parsing is done. Something like this:
public void parseWikitude(string url, Action callback) {
places = new List<POI>();
WebClient wc = new WebClient();
wc.DownloadStringCompleted += HttpsCompleted;
wc.DownloadStringAsync(new Uri(url), callback);
}
private void HttpsCompleted(object sender, DownloadStringCompletedEventArgs e)
{
if (e.Error == null)
{
...
var callback = (Action)e.UserState;
Deployment.Current.Dispatcher.BeginInvoke(callback);
}
}
//And then when you use it, you do it like that
parserXML.parseWikitude("http://myurl.php", delegate() {
//The code here will be executed on the UI thread, after the parsing is done
});
Use the (rather) new asnyc pattern in .NET. You should read about this, as it is one of the best features of .NET if you ask me. :) It basically does the callback thing automatically and makes the code a lot easier to read/maintain/work-with. Once you get used to it, that is.
Here's an example:
public Task<List<POI>> parseWikitude(string url) {
TaskCompletionSource<List<POI>> resultTaskSource = new TaskCompletionSource<List<POI>>();
WebClient wc = new WebClient();
wc.DownloadStringCompleted += HttpsCompleted;
wc.DownloadStringAsync(new Uri(url), resultTaskSource);
return resultTaskSource.Task;
}
private void HttpsCompleted(object sender, DownloadStringCompletedEventArgs e)
{
if (e.Error == null)
{
//If needed, run the code here in a background thread
//...
var resultTaskSource = (TaskCompletionSource<List<POI>>)e.UserState;
resultTaskSource.SetResult(places);
}
}
//And when you need to use it, do it like that (note, this must be invoked in an async method!)
var places = await parser.parseWikitude("http://myurl.php");
//The code here will be executed on the same thread when the parsing is done, but the thread will not be blocked while the download is happening.
So, these are the two ways you can handle it. Option one is old-school, classic and easy. Option two is the new and cool way of doing async stuff. It really is a must-know. Simplifies a lot of things once you get used to it.
P.S. Sorry if I got carried away. :D
I have a webrequest to get a xml.That works great but when i press F12(lock screen) while the the server is requested by my app...I got a WebException.
I use a taskCompeltionSource object...Here is my code
public async Task<String> Query(DataRequestParam dataRequestParam)
{
_dataRequestParam = dataRequestParam;
try
{
Result = "";
Result = await myDownloadString(dataRequestParam);
}
catch (WebException we)//ERROR IS CAUGHT HERE
{
throw new WebException(we.Message);
}
catch (Exception ex)
{
throw new MyException(ex.Message);
}
return Result;
}
public static Task<string> myDownloadString(DataRequestParam dataRequestParam)
{
var tcs = new TaskCompletionSource<string>();
var web = new WebClient();
if (!string.IsNullOrEmpty(dataRequestParam.AuthentificationLogin))
{
System.Net.NetworkCredential account = new NetworkCredential(dataRequestParam.AuthentificationLogin, dataRequestParam.AuthentificationPassword);
web.Credentials = account;
}
web.DownloadStringCompleted += (s, e) =>
{
if (e.Error != null) tcs.TrySetException(e.Error);
else if (e.Cancelled) tcs.TrySetCanceled();
else tcs.TrySetResult(e.Result);
};
web.DownloadStringAsync(dataRequestParam.TargetUri);
return tcs.Task;
}
If you haven't disabled ApplisationIdleDetection, your process is stopped while entering Lock screen - thus you probably get the exception - like I've said in comment. Disabling will solve this issue, but you must be aware of few things:
you will still get the exception when hitting Start Button (or other case putting your app to dormant state). In this case your app is stopped and there is no way to prevent this behaviour.
you must fulfill certification requirements when disabling App Idle Detection - point 6.3
if you want to download files in the Background (lock screen, after closing/leaving app) then you can think of Background Transfers
I am trying to implement jsr-179 APi into Nokia Symbian phone for periodic location update using setLocationListener through J2me. In emulator it is working fine. While I installed Midlet on the device nokia 5230, it is given NullPointerException and the application is automatically terminating. What might be possible causes?
Below is my class, I am instantiating object for this class on a form in netbeans
class MovementTracker implements LocationListener {
LocationProvider provider;
Location lastValidLocation;
UpdateHandler handler;
boolean done;
public MovementTracker() throws LocationException
{
done = false;
handler = new UpdateHandler();
new Thread(handler).start();
//Defining Criteria for Location Provider
/*
Criteria cr = new Criteria();
cr.setHorizontalAccuracy(500);
*/
//you can place cr inside getInstance
provider = LocationProvider.getInstance(null);
//listener,interval,timeout,int maxAge
//Passing -1 selects default interval
// provider.setLocationListener(MovementTracker.this, -1, -1, -1);
provider.setLocationListener(MovementTracker.this, -1, 30000, 30000);
}
public void locationUpdated(LocationProvider provider, Location location)
{
handler.handleUpdate(location);
batteryLevel = System.getProperty("com.nokia.mid.batterylevel");
sn = System.getProperty("com.nokia.mid.networksignal");
localTime = System.currentTimeMillis();
Send_Location();
}
public void providerStateChanged(LocationProvider provider, int newState)
{
}
class UpdateHandler implements Runnable
{
private Location updatedLocation = null;
// The run method performs the actual processing of the location
public void run()
{
Location locationToBeHandled = null;
while (!done)
{
synchronized(this)
{
if (updatedLocation == null)
{
try
{
wait();
}
catch (Exception e)
{
// Handle interruption
}
}
locationToBeHandled = updatedLocation;
updatedLocation = null;
}
// The benefit of the MessageListener is here.
// This thread could via similar triggers be
// handling other kind of events as well in
// addition to just receiving the location updates.
if (locationToBeHandled != null)
processUpdate(locationToBeHandled);
}
try
{
Thread.sleep(10000); //Sleeps for 10 sec & then sends the data
}
catch (InterruptedException ex)
{
}
}
public synchronized void handleUpdate(Location update)
{
updatedLocation = update;
notify();
}
private void processUpdate(Location update)
{
latitude = update.getQualifiedCoordinates().getLatitude();
longitude = update.getQualifiedCoordinates().getLongitude();
altitude = update.getQualifiedCoordinates().getAltitude();
}
}
}
public MovementTracker() throws LocationException
...
I have not written any code for handling LocationException.
No code is very dangerous practice, just search the web for something like "java swallow exceptions".
It is quite possible that because of implementation specifics Nokia throws LocationException where emulator does not throw it. Since you don't handle exception this may indeed crash you midlet at Nokia - and you wouldn't know the reason for that because, again, you have written no code to handle it.
How can I catch that exception?
The simplest thing you can do is to display an Alert with exception message and exit the midlet after user reads and dismisses alert
so when i run this code to try to change the background the GUI crashes and gets stuck in a infinite while loop ignoring the event listeners. here is the code:
private Panel getPanel1() {
if (panel1 == null) {
panel1 = new Panel();
panel1.setLayout(new GridBagLayout());
while(frame.isVisible()){
panel1.addMouseListener(new java.awt.event.MouseAdapter() {
public void mouseClicked(java.awt.event.MouseEvent e) {
frame.setVisible(false);
}
});
int r = (int) (Math.random()*255);
int g = (int) (Math.random()*255);
int b = (int) (Math.random()*255);
Color c = new Color(r, g, b);
panel1.setBackground(c);
try {
Thread.sleep(4000);
} catch (InterruptedException e1) {
e1.printStackTrace();
}
panel1.addMouseListener(new java.awt.event.MouseAdapter() {
public void mouseClicked(java.awt.event.MouseEvent e) {
/*panel1.setVisible(false);
frame.setVisible(false);*/
System.exit(0);
}
});
}
}
return panel1;
}
instead of exiting the loop of terminating the program or event changing the background it just displays the panel and does nothing else and i have to force it to quit. what should i do?
You're effectively blocking the UI thread by calling sleep in a loop. In that loop you're also adding two listeners on every iteration too, which is quite bizarre.
Don't block the UI thread. Let the GUI framework take care of delivering events etc. Basically you need to take an event-based approach to UI, rather than the approach you currently are taking, which will never let any events get despatched (as you're never returning control to the caller).
Create the panel, add the appropriate event listener, and then just return it to the caller. If you want to change the background colour every 4 seconds, you should do that via a timer so that it's not blocking the UI thread waiting for the 4 seconds to elapse.