danielbinder / DanielsCache

A simple caching library for keeping a certain fixed amount of your computed objects with an LRU eviction strategy

DanielsCache

A simple caching library for keeping a certain fixed amount of your computed objects with an LRU eviction strategy.

---

Time Complexity

Complexity (usually)	Description
O(mapper function)	Fetching unseen element
O(1)	Fetching seen element
O(1)	Eviction of LRU element

How this works internally

Fetching an unseen element

• The result is computed and the value is stored in the Hashmap in a new Node
• The element is added as last

Fetching a seen element

• The result is stored in the Hashmap, so it can be fetched and extracted from the Node
• The element is removed from its current position and prev and next get linked together
• The element is added as last

Eviction of the LRU element

• first points to the LRU element
• first.object is removed from the Hashmap
• first.next is the new first
• first.prev gets unlinked

Example of how it could look on the inside

I want to use and/or adapt this project

Go for it - I tried to make everything as readable and modifiable as possible. Check out the explanation below, and the license in the 'LICENSE' file. Regardless of the license, it would be cool if you somehow mentioned, that you got this code from here :)

How to use

1. Add as library (here shown in IntelliJ) or add the code directly to your project
2. Create a new Cache instance with DanielsCache<O, R> cache = new DanielsCache<>(int maxSize, Function<O, R> mapper);
   • O is the input type e.g. a String from an incoming REST request
   • R is the computed type e.g. the object created from the REST string
3. Use cache.get(O object) to get the corresponding R result
   • The result is computed if it is not in the cache
   • But simply fetched if it is
   • Eventually the cache fills up, and the LRU (least recently used) element is evicted