Question

I have ViewPager which I use to show zoomable images (using ImageViewTouch).
I need to load large bitmaps from Internet (http). By large I mean 2000×1000. The images need to be so large, because they are zoomable and need to show details. Images on server are .jpg format, but it’s not problem – I can change it.

How can I manage to load so large images to ImageViewTouch (ImageView) without getting probles with memory?

By now I’m using simply this (AsyncTask):

    ImageView currentView; //ImageView where to place image loaded from Internet
    String ImageUrl; //URL of image to load

    protected Bitmap doInBackground(ArrayList... params) {

            Bitmap bitmap;
            InputStream in = null;
            imageUrl = (String)params[0].get(1);

            try{
                HttpClient httpclient = new DefaultHttpClient();
                HttpResponse response = httpclient.execute(new HttpGet(imageUrl));
                in = response.getEntity().getContent();
            } catch(Exception e){
                e.printStackTrace();
            }
            try {
                bitmapa = BitmapFactory.decodeStream(in);
                in.close();
            } catch (IOException e1) {
                e1.printStackTrace();
            }

            return bitmap;
        }


        @Override
        protected void onPostExecute(Bitmap bitmap) {

            currentView.setImageBitmap(bitmap);
        }

And it cause many problems with memory:

E/dalvikvm-heap(369): 69560740-byte external allocation too large for this process.
E/GraphicsJNI(369): VM won't let us allocate 69560740 bytes

or

E/AndroidRuntime(369): java.lang.RuntimeException: An error occured while executing doInBackground()
E/AndroidRuntime(369):  at android.os.AsyncTask$3.done(AsyncTask.java:200)
E/AndroidRuntime(369):  at java.util.concurrent.FutureTask$Sync.innerSetException(FutureTask.java:274)
E/AndroidRuntime(369):  at java.util.concurrent.FutureTask.setException(FutureTask.java:125)
E/AndroidRuntime(369):  at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:308)
E/AndroidRuntime(369):  at java.util.concurrent.FutureTask.run(FutureTask.java:138)
E/AndroidRuntime(369):  at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1088)
E/AndroidRuntime(369):  at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:581)
E/AndroidRuntime(369):  at java.lang.Thread.run(Thread.java:1019)
E/AndroidRuntime(369): Caused by: java.lang.OutOfMemoryError: bitmap size exceeds VM budget
E/AndroidRuntime(369):  at android.graphics.BitmapFactory.nativeDecodeStream(Native Method)
E/AndroidRuntime(369):  at android.graphics.BitmapFactory.decodeStream(BitmapFactory.java:470)
E/AndroidRuntime(369):  at android.graphics.BitmapFactory.decodeStream(BitmapFactory.java:525)
E/AndroidRuntime(369):  at com.package.app.ImageDownloader.doInBackground(ImageDownloader.java:78)
E/AndroidRuntime(369):  at com.package.app.ImageDownloader.doInBackground(ImageDownloader.java:1)
E/AndroidRuntime(369):  at android.os.AsyncTask$2.call(AsyncTask.java:185)
E/AndroidRuntime(369):  at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:306)
E/AndroidRuntime(369):  ... 4 more

Answer 1

What version of the OS do you test this on? On earlier versions the heap space available is much lower than on later versions.

I would seriously consider downsizing your bitmaps to avoid this. If you are downloading a 2000×1000 Bitmap for an mdpi phone then this is probably a bad idea.

I would also recommend reading this article for more information on how to load the appropriate image exactly for a specific ImageView, based on its dimensions.

Finally, you should always close an InputStream in a finally block:

try {
        bitmap = BitmapFactory.decodeStream(in);
    } catch (IOException e) {
        e.printStackTrace();
    } finally {
        if (in != null) { in.close(); }
    }

There is also android:largeHeap="true" which you can consider, specifically made for apps that deal with large bitmaps, such as photo editing apps.