Cache - musings and implementation Currently working on UI design for my forum. In principle such as the simple solution is already ready. So I decided to get on with the slow implementations of the basic views. And here came the first problem. Namely, management, compression and caching JS and CSS files. As usual, we have that in the draft some solid scripts (eg jQuery), which are used for each view, so embarrassing that they were cached and compressed. So I sat down to solve this problem or implementations of the wrapper to the standard cache mechanism which is available in ASP.NET. Just as with HttpContextem I want to be highly touted.
Interface and Class Cache As you can guess the cache class will be transposed ICache interface. First, it presents what it looks like later a brief description. Public interface
ICache
{
\u0026lt;T> void Set (string key, object value, DateTime absoluteExpiration,
slidingExpiration TimeSpan);
\u0026lt;T> void Set (string key, object value, DateTime absoluteExpiration);
void Set & # 60, T> (string key, object value, TimeSpan slidingExpiration);
\u0026lt;T> T Remove (string key);
void Clear ();
Get \u0026lt;T> T ( string key);}
interface has several methods that set the value in the cache. They differ on variables of the Expiration of the cache. AbsoluteExpiration variable specifies an absolute expiration date of the object that is SlidingExpiration but does not specify an expiration date and the time interval after which the value is due to expire. The rest should not raise doubts. The actual implementation looks like this: public class
Cache: ICache
{/ / omit variable declarations and functions set
...
\u0026lt;T> public T Remove (string key) {
return (T) _cache.Remove (key);}
public void Clear () {
IDictionaryEnumerator _cache.GetEnumerator cacheEnumerator = () ;
while (cacheEnumerator.MoveNext())
{
_cache.Remove(cacheEnumerator.Key.ToString());
}
}
public T Get<T>(string key)
{
Check.Argument.IsNotEmpty(key, "key");
lock (_syncObject)
{
T cacheItem = (T)_cache.Get(key);
if (cacheItem == null)
return default(T);
else
return cacheItem;
}
}
private void DoSet<T>(string key, object value,
absoluteExpiration DateTime, TimeSpan slidingExpiration)
{
Check.Argument.IsNotEmpty (key, "key");
Check.Argument.IsNotNull (value, "value");
if (_cache [key] == null)
_cache.Add (key, value, null, absoluteExpiration, slidingExpiration, CacheItemPriority.Normal, null);
else
_cache.Insert (key, value, null, absoluteExpiration,
slidingExpiration);
}}
I put only the most important parts of the code, I omitted all versions fukcji set as they relate to a private function DoSet with different parameters. As you can see the methods here operate on a cache of HttpRuntime.Cache. Clear function retrieves an enumerator from the collection and removes all objects from the cache. Get The function retrieves the value from cache, if there is it returns the default value for the object. It is thread-safe (secure multi-threaded). The method DoSet inserted object to the cache. It checks whether there is an object with a given key. If so, instead of adding just inserts the new value to it.
In the next post should now show a mechanism related to the caching of CSS and JS files. Of course, I added tests associated with caching. In addition, slowly threw a graphical interface which will take effect in the forum.
