前些日子一直为图片内存溢出问题困扰着,查了N多资料,将google彻底翻遍了都没找到解决方案,就当我几乎绝望的时候意外发现了一位网友的一个工具类,抱着最后一丝希望将代码co过来试了一把,结果令我喜出望外。嘿,解决了!暂不说多么欢喜了,听我慢慢道来这其中的前因后果吧!
需求:下载时候将图片一并down下来,在空间里显示并支持离线观看
第一个版本代码:
//从本地读取图片
public Bitmap getBitmapFromSD(String filename) {
FileInputStream fi = null;
BufferedInputStream bi = null;
Bitmap bp = null;
try {
fi = new FileInputStream(filename);
bi = new BufferedInputStream(fi);
bp = BitmapFactory.decodeStream(bi);
} catch (IOException e) {
bp = null;
} finally {
try {
if (bi != null) {
bi.close();
}
if (fi != null) {
fi.close();
}
} catch (IOException e) {
bp = null;
}
}
return bp;
}
问题出现了,由于显示的图片过大,所以会出现OutOfMemoryException。我就设想能否捕捉异常来回收图片再重新加载,于是欲从网上找解决办法,什么手动干预GC,什么将图片弱化什么使用弱引用保存图片,有些总结得特别好(http://mzh3344258.blog.51cto.com/1823534/804237),这些方法我一一尝试可问题仍然未解决。不断的OOM,不断的尝试recycle,错误倒是不出现,可一旦内存吃不消就会显示不了图片,出现的都是默认图片。最终我从网上找到如下工具类,助我很好的解决了此问题,具体网址忘记了(得谢谢那位网友啦(*^__^*) ),现在代码贴出来以便下次顺手拈来
public final class BitMapUtil {
private static final Size ZERO_SIZE = new Size(0, 0);
private static final Options OPTIONS_GET_SIZE = new Options();
private static final Options OPTIONS_DECODE = new Options();
private static final byte[] LOCKED = new byte[0];
// 此对象用来保持Bitmap的回收顺序,保证最后使用的图片被回收
private static final LinkedList CACHE_ENTRIES = new LinkedList();
// 线程请求创建图片的队列
private static final Queue TASK_QUEUE = new LinkedList();
// 保存队列中正在处理的图片的key,有效防止重复添加到请求创建队列
private static final Set TASK_QUEUE_INDEX = new HashSet();
// 缓存Bitmap
private static final Map IMG_CACHE_INDEX = new HashMap(); // 通过图片路径,图片大小
private static int CACHE_SIZE = 20; // 缓存图片数量
static {
OPTIONS_GET_SIZE.inJustDecodeBounds = true;
// 初始化创建图片线程,并等待处理
new Thread() {
{
setDaemon(true);
}
public void run() {
while (true) {
synchronized (TASK_QUEUE) {
if (TASK_QUEUE.isEmpty()) {
try {
TASK_QUEUE.wait();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
}
QueueEntry entry = TASK_QUEUE.poll();
String key = createKey(entry.path, entry.width,
entry.height);
TASK_QUEUE_INDEX.remove(key);
createBitmap(entry.path, entry.width, entry.height);
}
}
}.start();
}
public static Bitmap getBitmap(String path, int width, int height) {
if(path==null){
return null;
}
Bitmap bitMap = null;
try {
if (CACHE_ENTRIES.size() >= CACHE_SIZE) {
destoryLast();
}
bitMap = useBitmap(path, width, height);
if (bitMap != null && !bitMap.isRecycled()) {
return bitMap;
}
bitMap = createBitmap(path, width, height);
String key = createKey(path, width, height);
synchronized (LOCKED) {
IMG_CACHE_INDEX.put(key, bitMap);
CACHE_ENTRIES.addFirst(key);
}
} catch (OutOfMemoryError err) {
destoryLast();
System.out.println(CACHE_SIZE);
return createBitmap(path, width, height);
}
return bitMap;
}
public static Size getBitMapSize(String path) {
File file = new File(path);
if (file.exists()) {
InputStream in = null;
try {
in = new FileInputStream(file);
BitmapFactory.decodeStream(in, null, OPTIONS_GET_SIZE);
return new Size(OPTIONS_GET_SIZE.outWidth,
OPTIONS_GET_SIZE.outHeight);
} catch (FileNotFoundException e) {
return ZERO_SIZE;
} finally {
closeInputStream(in);
}
}
return ZERO_SIZE;
}
// ------------------------------------------------------------------ private Methods
// 将图片加入队列头
private static Bitmap useBitmap(String path, int width, int height) {
Bitmap bitMap = null;
String key = createKey(path, width, height);
synchronized (LOCKED) {
bitMap = IMG_CACHE_INDEX.get(key);
if (null != bitMap) {
if (CACHE_ENTRIES.remove(key)) {
CACHE_ENTRIES.addFirst(key);
}
}
}
return bitMap;
}
// 回收最后一张图片
private static void destoryLast() {
synchronized (LOCKED) {
String key = CACHE_ENTRIES.removeLast();
if (key.length() > 0) {
Bitmap bitMap = IMG_CACHE_INDEX.remove(key);
if (bitMap != null && !bitMap.isRecycled()) {
bitMap.recycle();
bitMap = null;
}
}
}
}
// 创建键
private static String createKey(String path, int width, int height) {
if (null == path || path.length() == 0) {
return "";
}
return path + "_" + width + "_" + height;
}
// 通过图片路径,宽度高度创建一个Bitmap对象
private static Bitmap createBitmap(String path, int width, int height) {
File file = new File(path);
if (file.exists()) {
InputStream in = null;
try {
in = new FileInputStream(file);
Size size = getBitMapSize(path);
if (size.equals(ZERO_SIZE)) {
return null;
}
int scale = 1;
int a = size.getWidth() / width;
int b = size.getHeight() / height;
scale = Math.max(a, b);
synchronized (OPTIONS_DECODE) {
OPTIONS_DECODE.inSampleSize = scale;
Bitmap bitMap = BitmapFactory.decodeStream(in, null,
OPTIONS_DECODE);
return bitMap;
}
} catch (FileNotFoundException e) {
Log.v("BitMapUtil","createBitmap=="+e.toString());
} finally {
closeInputStream(in);
}
}
return null;
}
// 关闭输入流
private static void closeInputStream(InputStream in) {
if (null != in) {
try {
in.close();
} catch (IOException e) {
Log.v("BitMapUtil","closeInputStream=="+e.toString());
}
}
}
// 图片大小
static class Size {
private int width, height;
Size(int width, int height) {
this.width = width;
this.height = height;
}
public int getWidth() {
return width;
}
public int getHeight() {
return height;
}
}
// 队列缓存参数对象
static class QueueEntry {
public String path;
public int width;
public int height;
}
}
在使用时我只调用了getBitmap方法,将需要设置的高度宽度以及本地图片路径传递过去就能自动返回bitmap给我,而且当捕捉到OOMError的时候将LinkedList的最后一张图片也就是最先存的图片进行溢出并回收就大功告成,特别注意的是这里捕捉错误Exception是获取不到的,一定要手动捕获OutOfMemoryError你才能进行处理(估计这些道理大家都懂得,所以不赘述啦,童鞋们加油!办法总比困难多o(∩_∩)o )