Redis using Spring Data – Part 2/3

This is the second part of tutorial for Redis using Spring Data series. In the first part we have covered installing redis server, using redis cli and two minute tutorial about redis data types. In case you have missed it you can revisit the first part of the tutorial.

In this article we will create a java application using step by step approach using maven and spring-data-redis to explore various data structures supported by redis. This article assumes that you have working installation of maven on your system.

Create the Project
Run the following command from your terminal to create a template java project using maven.
$ mvn archetype:generate -DgroupId=com.wordpress.omanandj -DartifactId=spring-data-redis-demo -DarchetypeArtifactId=maven-archetype-quickstart -DinteractiveMode=false

This will generate the basic project structure in the directory spring-data-redis-demo. Open the pom.xml file and replace the content with following:

Spring beans configuration
In this section we will use java based configuration to plugin spring beans for our demo project. We will use redis recommended java client library jedis for communicating with redis database. Spring-data-redis provides RedisTemplate. For most basic use case we will use string focused extension of Redis Template – StringRedisTemplate.
Create a class AppConfiguration in the package com.wordpress.omanandj.config to plugin spring beans. The AppConfiguration class content can be copied from following content :

Create a simple String based Key/Value cache implementation using Redis
Let’s create a persistent distributed caching storage using redis. Our cache repository interface has basic api requirement to be able to store a single key/value in datastore, multiple key/values in datastore in a single network call, retrieve a single key or multiple key, remove a key. Create an generic interface ICacheRepository for our API requirement as:

Now create a simple cache repository implementation for the interface ICacheRepository with string as key and value to explore . The implementation of saving an key/value can be done as:

Here opsForValue method provides the interface for saving simple key/value pair in the redis datastore. The set method is equivalent to $ set key value. Now implement other methods of the ICacheRepository as:

Now let us test our SimpleCacheRepository implementation. Open the App.java in the project and test the code by pasting following:

You should verify your results in the redis datastore using redis-cli.

Using Redis List Data Type
For operations on the list data type Redistemplate gives opsForList() method. Redis list support pushing an element to the left and right, popping an element from left and right, getting elements of the list, trimming the list etc. Based on these requirements create a demo interface for showcasing each of these functionality:

The IRedisListDemo can be implemented as :

Now let us test our RedisListTypeRepository by adding following code snippet into the App.java:

As you may have noticed implementation of the multiAdd is not optimal and I still need to find the same with RedisTemplate and Jedis client. However, redis-cli does support it.

$ LPUSH marks 10 20 30 40
(integer) 4
$ lrange marks 0 3
1) "40"
2) "30"
3) "20"
4) "10"


Using Redis Set Data Type
As described in first post redis set data type is an unordered collection of unique elements. Similar to list following classes demonstrate the adding an element to the set, getting members of the set, verify if a value is the member of the set.

The sample implementation of the IRedisSetDemo is defined in RedisSetTypeRepository as follows:

Add following skeleton of the code to test RedisSetTypeRepository:

Using Redis Hashes Data Type
Redis Hashes are maps between string fields and string values. Redis Hashes are most suitable to represent objects. As per redis documentation hash with nearly 100 fields are stpred in a way that takes very little space, so that you can store millions of objects in a small Redis instance.
For demonstrating the hash data structure for redis, let’s create a model object User with fields- id, name, age, organization and password as:

Now create an Interface IHashRepository for describing of the redis commands for saving and retrieving objects from the redis datastore as:

Now let us implement the interface IHashRepository for user model object as UserRepository :

You may have noticed the difference of bean wiring for redisTemplate in UserRepository. Since the UserRepositry not uses String as values, we have to define a new bean for redisTemplate. Let us define a more generic bean in your AppConfiguration class as:

Now let us test the UserRepository by adding following code skelton to your App.java:

Thats it for this second part of this series. I hope this will help you more deeper into the thought of redis and spring-data-redis for different data structure.
We will continue developing our spring-data-redis-demo app in the third part of this series to understand the Publish subscribe messaging by Redis PubSub using spring. May be we will have a 2 minute introduction for the replication setup for redis.