Hey all, I've been spending a lot of time in the Camera code recently trying to improve it in a number of ways. First off the way we were loading the entire image into a Bitmap (stored in memory) was causing some problems. For instance on my Samsung Galaxy Note a picture takes up 30 mb of memory when loaded into a Bitmap. This was causing out of memory errors in apps that use the camera.
I've gone through the code and made a number of improvements. 1) If the call to get picture does not require a width, height or quality change we no longer load the image into a Bitmap as it is unnecessary. Saves a ton of memory. 2) If the call to get picture requires a width and height change we no longer load the entire image into a Bitmap. First we take advantage of the input sample size to load a smaller version of the image which is closer to the width and height of the one requested by the user. For instance if users calls get picture asking for a 900 x 900 image we load the image into memory using a sample size of 2 which gives us a 1000 x 1000 image that is resized to 900 x 900. This saves a bunch of unnecessary memory usage as well as a 2000x2000 image is 15.6 mb while a 1000x1000 image is only 3.9 mb so the benefit on memory usage is obvious. 3) I fixed an issue on Samsung devices where a picture is still stored in the Gallery even if you set saveToPhotoAlbum to false. 4) I fixed the long standing issue where a picture taken in portrait mode ends up being displayed on it's side in the webview. This is because the web view just doesn't care about exif orientation information. Now I rotate all images to the normal orientation. Images display correctly in the webview, Gallery and on the desktop. So I really would like to check this in before 1.9.0 is release but I'd love if someone could look over the code just to make sure my changes are sane. https://github.com/macdonst/cordova-android/tree/cam2 Thanks... Simon Mac Donald http://hi.im/simonmacdonald
