# The Redis proxy
## Requirement
For the assignment the requirements are listed below
### Web Service
* There is a web server listening port configured by config file: resources/config.properties
* The web server only takes query parameters as this is a toy project
* for example ?key={....}&value={...} for POST (write)
* And ?key={...} for GET (read)
### Backend with Redis
* The proxy is configured to talk to redis through the config.properties file mentioned above
### Local Cache: TTLLRUCache: (aka: TTL + LRU cache)
* Use specified key to store cache
* Global expiry time specified on the global configuration
* Capacity is specified on the global configuration in terms of the number of keys
* LRU eviction and TTL expiry for the cache keys
* The implementation is based on Twitter's storehaus's TTL and LRU cache combined
### Global Configuration
* Address and port of the backend Redis
* Expiry time
* Capacity
* Port number of listening
* Max concurrency
### Concurrency
* Client should be able to connect concurrently
* Internally requests are able to process sequentially
* The implementation here utilize finagle libraries' maxConcurrentRequests.
### Tests
* End to end system function tests
* The system tests need a backing redis server as there is no suitable embedded redis server implementation
* Proxy should connect to a running Redis instance
### How to local basic test
* Start a local redis service, `redis-server /usr/local/etc/redis.conf`
* Start this redis-proxy service `./sbt 'runMain com.github.bpin.Main'`
* Use a curl command:
* GET: curl "localhost:9090/?key=joe"
* POST: curl -X POST -D - "localhost:9090/?key=name&value=brian"