java lrucache_基于Java实现缓存Cache的深入分析

原理是使用LinkedHashMap来实现，当缓存超过大小时，将会删除最老的一个元组。

实现代码如下所示

import java.util.LinkedHashMap;

import java.util.Map;

public class LRUCache {

public static class CachedData {

private Object data = null;

private long time = 0;

private boolean refreshing = false;

public CachedData(Object data) {

this.data = data;

this.time = System.currentTimeMillis();

}

public Object getData() {

return data;

}

public long getTime() {

return time;

}

public void setTime(long time) {

this.time = time;

}

public boolean getRefreshing() {

return refreshing;

}

public void setRefreshing(boolean b) {

this.refreshing = b;

}

}

protected static class CacheMap extends LinkedHashMap {

protected int maxsize = 0;

public CacheMap(int maxsize) {

super(maxsize * 4 / 3 + 1, 0.75f, true);

this.maxsize = maxsize;

}

protected boolean removeEldestEntry(Map.Entry eldest) {

return size() > this.maxsize;

}

}

protected CacheMap map = null;

public LRUCache(int size) {

this.map = new CacheMap(size);

}

public synchronized void set(Object key, Object value) {

map.remove(key);

map.put(key, new CachedData(value));

}

public synchronized void remove(Object key) {

map.remove(key);

}

public synchronized CachedData get(Object key) {

CachedData value = (CachedData) map.get(key);

if (value == null) {

return null;

}

map.remove(key);

map.put(key, value);

return value;

}

public int usage() {

return map.size();

}

public int capacity() {

return map.maxsize;

}

public void clear() {

map.clear();

}

}