不得不说,用哈希操作来存对象,有点自讨苦吃!
不过,既然吃了苦,也做个记录,也许以后API升级后,能好用些呢?!
或许,是我的理解不对,没有真正的理解哈希表。
相关链接:
Redis实战之征服 Redis + Jedis + Spring (一)
Redis实战之征服 Redis + Jedis + Spring (二)
Redis实战之征服 Redis + Jedis + Spring (三)
一、预期
接上一篇,扩充User属性:
- public class User implements Serializable {
- private static final long serialVersionUID = -1267719235225203410L;
- private String uid;
- private String address;
- private String mobile;
- private String postCode;
- }
public class User implements Serializable {
private static final long serialVersionUID = -1267719235225203410L; private String uid; private String address; private String mobile; private String postCode;
}
我期望的是:
redis 127.0.0.1:6379> hmget uc.user.info.uid.u123456 address mobile postCode
1) "\xe4\xb8\x8a\xe6\xb5\xb7"
2) "13800138000"
3) "100859"
1) "\xe4\xb8\x8a\xe6\xb5\xb7"
2) "13800138000"
3) "100859"
几乎就是一个对象了!
但是,接下来的代码实现,让我彻底崩溃了!
二、代码实现
1.保存——HMSET
- @Override
- public void save(final User user) {
- redisTemplate.execute(new RedisCallback<Object>() {
- @Override
- public Object doInRedis(RedisConnection connection)
- throws DataAccessException {
- byte[] key = redisTemplate.getStringSerializer().serialize(
- "uc.user.info.uid." + user.getUid());
- BoundHashOperations<Serializable, byte[], byte[]> boundHashOperations = redisTemplate
- .boundHashOps(key);
- boundHashOperations.put(redisTemplate.getStringSerializer()
- .serialize("mobile"), redisTemplate
- .getStringSerializer().serialize(user.getMobile()));
- boundHashOperations.put(redisTemplate.getStringSerializer()
- .serialize("address"), redisTemplate
- .getStringSerializer().serialize(user.getAddress()));
- boundHashOperations.put(redisTemplate.getStringSerializer()
- .serialize("postCode"), redisTemplate
- .getStringSerializer().serialize(user.getPostCode()));
- connection.hMSet(key, boundHashOperations.entries());
- return null;
- }
- });
- }
@Override
public void save(final User user) {
redisTemplate.execute(new RedisCallback<Object>() {
@Override
public Object doInRedis(RedisConnection connection)
throws DataAccessException {
byte[] key = redisTemplate.getStringSerializer().serialize(
"uc.user.info.uid." + user.getUid());
BoundHashOperations<Serializable, byte[], byte[]> boundHashOperations = redisTemplate
.boundHashOps(key);
boundHashOperations.put(redisTemplate.getStringSerializer()
.serialize("mobile"), redisTemplate
.getStringSerializer().serialize(user.getMobile()));
boundHashOperations.put(redisTemplate.getStringSerializer()
.serialize("address"), redisTemplate
.getStringSerializer().serialize(user.getAddress()));
boundHashOperations.put(redisTemplate.getStringSerializer()
.serialize("postCode"), redisTemplate
.getStringSerializer().serialize(user.getPostCode()));
connection.hMSet(key, boundHashOperations.entries());
return null;
}
});
}
这里用到:
- BoundHashOperations<Serializable, byte[], byte[]> boundHashOperations = redisTemplate.boundHashOps(key);
- boundHashOperations.put(redisTemplate.getStringSerializer().serialize("mobile"), redisTemplate.getStringSerializer().serialize(user.getMobile()));
BoundHashOperations<Serializable, byte[], byte[]> boundHashOperations = redisTemplate.boundHashOps(key);
boundHashOperations.put(redisTemplate.getStringSerializer().serialize("mobile"), redisTemplate.getStringSerializer().serialize(user.getMobile()));
看着就有点肿。。。Map封装完以后,用HMSET命令:
- connection.hMSet(key, boundHashOperations.entries());
connection.hMSet(key, boundHashOperations.entries());
这时候就完成了哈希表的保存操作,可以在控制台看到相应的数据了。
redis 127.0.0.1:6379> hmget uc.user.info.uid.u123456 address mobile postCode
1) "\xe4\xb8\x8a\xe6\xb5\xb7"
2) "13800138000"
3) "100859"
1) "\xe4\xb8\x8a\xe6\xb5\xb7"
2) "13800138000"
3) "100859"
2.获取——HMGET
这一刻,我彻底崩溃了!取出来的值是个List,还得根据取得顺序,逐个反序列化,得到内容。
- @Override
- public User read(final String uid) {
- return redisTemplate.execute(new RedisCallback<User>() {
- @Override
- public User doInRedis(RedisConnection connection)
- throws DataAccessException {
- byte[] key = redisTemplate.getStringSerializer().serialize(
- "uc.user.info.uid." + uid);
- if (connection.exists(key)) {
- List<byte[]> value = connection.hMGet(
- key,
- redisTemplate.getStringSerializer().serialize(
- "address"),
- redisTemplate.getStringSerializer().serialize(
- "mobile"), redisTemplate
- .getStringSerializer()
- .serialize("postCode"));
- User user = new User();
- String address = redisTemplate.getStringSerializer()
- .deserialize(value.get(0));
- user.setAddress(address);
- String mobile = redisTemplate.getStringSerializer()
- .deserialize(value.get(1));
- user.setMobile(mobile);
- String postCode = redisTemplate.getStringSerializer()
- .deserialize(value.get(2));
- user.setPostCode(postCode);
- user.setUid(uid);
- return user;
- }
- return null;
- }
- });
- }
@Override
public User read(final String uid) {
return redisTemplate.execute(new RedisCallback<User>() {
@Override
public User doInRedis(RedisConnection connection)
throws DataAccessException {
byte[] key = redisTemplate.getStringSerializer().serialize(
"uc.user.info.uid." + uid);
if (connection.exists(key)) {
List<byte[]> value = connection.hMGet(
key,
redisTemplate.getStringSerializer().serialize(
"address"),
redisTemplate.getStringSerializer().serialize(
"mobile"), redisTemplate
.getStringSerializer()
.serialize("postCode"));
User user = new User();
String address = redisTemplate.getStringSerializer()
.deserialize(value.get(0));
user.setAddress(address);
String mobile = redisTemplate.getStringSerializer()
.deserialize(value.get(1));
user.setMobile(mobile);
String postCode = redisTemplate.getStringSerializer()
.deserialize(value.get(2));
user.setPostCode(postCode);
user.setUid(uid); return user;
}
return null;
}
});
}
这个实现,跟Redis的命令几乎一模一样,指定Key,指定field,获取其值。
- List<byte[]> value = connection.hMGet(key,redisTemplate.getStringSerializer().serialize("address"),
- redisTemplate.getStringSerializer().serialize("mobile"),
- redisTemplate.getStringSerializer().serialize("postCode"));
List<byte[]> value = connection.hMGet(key,redisTemplate.getStringSerializer().serialize("address"),
redisTemplate.getStringSerializer().serialize("mobile"),
redisTemplate.getStringSerializer().serialize("postCode"));
我绝对相信,要么是我用的过于肤浅,低估了Spring的封装能力。或者,我该直接Json!等等,这不是MongoDB干的事情吗?!
PS:这两篇博客里操作的数据类型,只能是String类型,还没搞定除此以外任何类型。吾将上下而求索~~~
上述操作也许你吐了,接下来的代码,就再吐一次吧!
封装对象的时候,一定要记得次序。。。。这绝对不是一个优质代码的实现风格!
- User user = new User();
- String address = redisTemplate.getStringSerializer().deserialize(value.get(0));
- user.setAddress(address);
- String mobile = redisTemplate.getStringSerializer().deserialize(value.get(1));
- user.setMobile(mobile);
- String postCode = redisTemplate.getStringSerializer().deserialize(value.get(2));
- user.setPostCode(postCode);
User user = new User();
String address = redisTemplate.getStringSerializer().deserialize(value.get(0));
user.setAddress(address);
String mobile = redisTemplate.getStringSerializer().deserialize(value.get(1));
user.setMobile(mobile);
String postCode = redisTemplate.getStringSerializer().deserialize(value.get(2));
user.setPostCode(postCode);
好吧!苦逼的事情,就此结束。目标Json支持!
相关链接:
Redis实战之征服 Redis + Jedis + Spring (一)