SpringBoot整合Redis + SpringCache + Protobuf，优雅地实现key-value键值对存储DEMO。

提要

使用Protostuff作为序列化机制，它的特点是：基于Protobuf、高效的编解码性能、以及和其他序列化框架相比Protostuff在序列化后字节码流更小，更易于传输及存储。 接下来进入正文…

版本

jdk8， SpringBoot 2.1.9.RELEASE

项目源码

本文所用源码已经上传github。

核心类库导入

Redis相关类库

<!-- apache 连接池 --> <dependency> <groupId>org.apache.commons</groupId> <artifactId>commons-pool2</artifactId> </dependency> <!-- Redis --> <dependency> <groupId>org.springframework.boot</groupId> <artifactId>spring-boot-starter-data-redis</artifactId> </dependency>

Protostuff 类库

<!-- Protostuff 序列化&反序列化工具 --> <dependency> <groupId>io.protostuff</groupId> <artifactId>protostuff-core</artifactId> <version>1.6.0</version> </dependency> <dependency> <groupId>io.protostuff</groupId> <artifactId>protostuff-runtime</artifactId> <version>1.6.0</version> </dependency>

Redis核心配置文件

spring: cache: type: redis redis: time-to-live: ${REDIS_CACHE_TTL:1800000} redis: host: ${REDIS_CACHE_HOST:www.yorozuyas.com} port: ${REDIS_CACHE_PORT:6379} password: ${REDIS_CACHE_PASSWORD:password}

这里只做了简单的配置，其他相关配置自行查看源码或相关资料进行更改。

定义自己的序列化器

透过源码，可以发现每个序列化器都是继承自org.springframework.data.redis.serializer.RedisSerializer<T>接口

package org.springframework.data.redis.serializer; import org.springframework.lang.Nullable; ... public interface RedisSerializer<T> { /** * Serialize the given object to binary data. */ @Nullable byte[] serialize(@Nullable T t) throws SerializationException; /** * Deserialize an object from the given binary data. */ @Nullable T deserialize(@Nullable byte[] bytes) throws SerializationException; ... }

因此只需要自定义一个继承自该接口的类，并实现serialize、deserialize即可。

package com.yorozuyas.demo.cache; import org.springframework.data.redis.serializer.RedisSerializer; import org.springframework.data.redis.serializer.SerializationException; import io.protostuff.LinkedBuffer; import io.protostuff.ProtostuffIOUtil; import io.protostuff.Schema; import io.protostuff.runtime.RuntimeSchema; import lombok.AllArgsConstructor; import lombok.Getter; import lombok.NoArgsConstructor; /** * 使用Protostuff对value进行编解码。 */ public final class ProtostuffRedisSerializer implements RedisSerializer<Object> { final Schema<CacheValueWrapper> schema = RuntimeSchema.getSchema( CacheValueWrapper.class ); @Override public byte[] serialize( Object cvw ) throws SerializationException { try { return encode( cvw ); } catch ( RuntimeException e ) { throw new SerializationException( "Could not encode CacheValueWrapper to Protostuff: " + e.getMessage(), e ); } } @Override public Object deserialize( byte[] bytes ) throws SerializationException { try { return decode( bytes ); } catch ( RuntimeException e ) { throw new SerializationException( "Could not decode Protostuff to CacheValueWrapper: " + e.getMessage(), e ); } } // do serialize public byte[] encode( Object value ) { final LinkedBuffer buffer = LinkedBuffer.allocate(); return ProtostuffIOUtil.toByteArray( new CacheValueWrapper( value ), schema, buffer ); } // do deserialize public Object decode( byte[] bytes ) { CacheValueWrapper wrapper = new CacheValueWrapper(); ProtostuffIOUtil.mergeFrom( bytes, wrapper, schema ); return wrapper.getData(); } @AllArgsConstructor @NoArgsConstructor public static class CacheValueWrapper { @Getter private Object data; } }

如何替换Redis默认的序列化器

在源码org.springframework.boot.autoconfigure.cache.RedisCacheConfiguration这个配置类中，可以看到注入了一个类型为RedisCacheManager的Bean，RedisCacheManager在创建过程中，调用determineConfiguration方法，并判断如果已经存在 org.springframework.data.redis.cache.RedisCacheConfiguration这个Bean，则直接返回，否则创建一个默认的org.springframework.data.redis.cache.RedisCacheConfiguration交给RedisCacheManager.

package org.springframework.boot.autoconfigure.cache; ... @Configuration @ConditionalOnClass(RedisConnectionFactory.class) @AutoConfigureAfter(RedisAutoConfiguration.class) @ConditionalOnBean(RedisConnectionFactory.class) @ConditionalOnMissingBean(CacheManager.class) @Conditional(CacheCondition.class) class RedisCacheConfiguration { private final CacheProperties cacheProperties; private final CacheManagerCustomizers customizerInvoker; private final org.springframework.data.redis.cache.RedisCacheConfiguration redisCacheConfiguration; RedisCacheConfiguration(CacheProperties cacheProperties, CacheManagerCustomizers customizerInvoker, ObjectProvider<org.springframework.data.redis.cache.RedisCacheConfiguration> redisCacheConfiguration) { this.cacheProperties = cacheProperties; this.customizerInvoker = customizerInvoker; this.redisCacheConfiguration = redisCacheConfiguration.getIfAvailable(); } @Bean public RedisCacheManager cacheManager(RedisConnectionFactory redisConnectionFactory, ResourceLoader resourceLoader) { RedisCacheManagerBuilder builder = RedisCacheManager.builder(redisConnectionFactory) .cacheDefaults(determineConfiguration(resourceLoader.getClassLoader())); List<String> cacheNames = this.cacheProperties.getCacheNames(); if (!cacheNames.isEmpty()) { builder.initialCacheNames(new LinkedHashSet<>(cacheNames)); } return this.customizerInvoker.customize(builder.build()); } private org.springframework.data.redis.cache.RedisCacheConfiguration determineConfiguration( ClassLoader classLoader) { if (this.redisCacheConfiguration != null) { return this.redisCacheConfiguration; } Redis redisProperties = this.cacheProperties.getRedis(); org.springframework.data.redis.cache.RedisCacheConfiguration config = org.springframework.data.redis.cache.RedisCacheConfiguration .defaultCacheConfig(); config = config.serializeValuesWith( SerializationPair.fromSerializer(new JdkSerializationRedisSerializer(classLoader))); if (redisProperties.getTimeToLive() != null) { config = config.entryTtl(redisProperties.getTimeToLive()); } if (redisProperties.getKeyPrefix() != null) { config = config.prefixKeysWith(redisProperties.getKeyPrefix()); } if (!redisProperties.isCacheNullValues()) { config = config.disableCachingNullValues(); } if (!redisProperties.isUseKeyPrefix()) { config = config.disableKeyPrefix(); } return config; } }

由于RedisCacheManager负责创建和管理各个RedisCache，并为各个RedisCache提供默认配置，包含序列化器，过期时间等。

package org.springframework.data.redis.cache; ... public class RedisCacheManager extends AbstractTransactionSupportingCacheManager { ... /** * Configuration hook for creating {@link RedisCache} with given name and {@code cacheConfig}. */ protected RedisCache createRedisCache(String name, @Nullable RedisCacheConfiguration cacheConfig) { return new RedisCache(name, cacheWriter, cacheConfig != null ? cacheConfig : defaultCacheConfig); } ... }

因此我们只需要自定义一个org.springframework.data.redis.cache.RedisCacheConfiguration的Bean并交由Spring容器即可，后续在启动过程中，RedisCacheManager在创建的时候，会拿到我们注入到Spring中的RedisCacheConfiguration。

package com.yorozuyas.demo.cache; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.boot.autoconfigure.AutoConfigureAfter; import org.springframework.boot.autoconfigure.cache.CacheAutoConfiguration; import org.springframework.boot.autoconfigure.cache.CacheProperties; import org.springframework.boot.autoconfigure.cache.CacheProperties.Redis; import org.springframework.boot.autoconfigure.condition.ConditionalOnProperty; import org.springframework.context.annotation.Bean; import org.springframework.context.annotation.Configuration; import org.springframework.data.redis.cache.RedisCacheConfiguration; import org.springframework.data.redis.serializer.RedisSerializationContext.SerializationPair; @Configuration @AutoConfigureAfter(CacheAutoConfiguration.class) @ConditionalOnProperty(name = "spring.cache.type", havingValue = "redis", matchIfMissing = false) public class RedisCacheAutoConfiguration { @Autowired private CacheProperties cacheProperties; /** * Custom {@link org.springframework.data.redis.cache.RedisCacheConfiguration} * * * @see org.springframework.boot.autoconfigure.cache.RedisCacheConfiguration */ @Bean public RedisCacheConfiguration determineConfiguration() { final Redis redisProperties = cacheProperties.getRedis(); // Only open TTL, others close. RedisCacheConfiguration config = RedisCacheConfiguration.defaultCacheConfig() .disableCachingNullValues() .disableKeyPrefix() // here, add custom ProtostuffRedisSerializer for value. .serializeValuesWith( SerializationPair.fromSerializer( new ProtostuffRedisSerializer() ) ); if ( redisProperties.getTimeToLive() != null ) { config = config.entryTtl( redisProperties.getTimeToLive() ); } return config; } }

模拟场景

在核心类编写完成后，我们来设计一个场景，并将其运用起来。数据库：MySQL 5.7

CRUD收货地址场景

通常我们在买东西的时候，都需要一个明确的收件地址，而一个用户可以有多个收货地址。

收货地址数据库设计

表名：tb_deliver_address Nofielddescriptiondata typenot null1id主键biginty2uid用户唯一标识varchary3name收货人varchary4address收货地址varchary5mobile联系方式biginty 初始化 sql DROP TABLE IF EXISTS tb_deliver_address CASCADE; CREATE TABLE tb_deliver_address ( id BIGINT NOT NULL AUTO_INCREMENT COMMENT '主键id', uid varchar(32) NOT NULL COMMENT '用户唯一标识', name varchar(24) NOT NULL COMMENT '收货人', address varchar(200) NOT NULL COMMENT '收货地址', mobile BIGINT NOT NULL COMMENT '联系方式', PRIMARY KEY (id) ) ENGINE=InnoDB DEFAULT CHARSET=utf8; INSERT INTO tb_deliver_address(uid, name, address, mobile) VALUES ('zhangsan', '我的姐姐', '湖北省武汉市江汉区发展大道185号', 12345678910); INSERT INTO tb_deliver_address(uid, name, address, mobile) VALUES ('zhangsan', '我的妹妹', '安徽省合肥市经济技术开发区芙蓉路678号', 12345678911); INSERT INTO tb_deliver_address(uid, name, address, mobile) VALUES ('zhangsan', '我的弟弟', '浙江省杭州市滨江区江南大道龙湖滨江天街', 12345678912); INSERT INTO tb_deliver_address(uid, name, address, mobile) VALUES ('zhangsan', '张三', '广东省佛山市顺德区大良迎宾路碧桂园·钻石湾', 12345678913); mapper POJO package com.yorozuyas.demo.model; import lombok.Getter; import lombok.Setter; import lombok.ToString; @Getter @Setter @ToString public class DeliverAddress { private Long id; private String uid; private String name; private String address; private Long mobile; }

API设计及实现

Controller package com.yorozuyas.demo.controller; ... import com.yorozuyas.demo.controller.base.BaseResponseEntity; import com.yorozuyas.demo.enums.Code; import com.yorozuyas.demo.model.DeliverAddress; import com.yorozuyas.demo.service.Protobuf2RedisService; import lombok.extern.slf4j.Slf4j; @RestController @Slf4j @RequestMapping(value = "/address") public class Protobuf2RedisController { public static final String URL_NEW_ADDRESS = "/new"; public static final String URL_FETCH_ADDRESS = "/fetch"; public static final String URL_MODIFY_ADDRESS = "/modify"; public static final String URL_EVICT_ADDRESS = "/evict/{id}"; @Autowired private Protobuf2RedisService protobuf2RedisService; ... private void checkIn( DeliverAddress addressInfo ) { if ( StringUtils.isNullOrEmpty( addressInfo.getUid() ) ) throw HttpClientErrorException.create( HttpStatus.BAD_REQUEST, "'uid' must not be null.", null, null, StandardCharsets.UTF_8 ); if ( StringUtils.isNullOrEmpty( addressInfo.getName() ) ) throw HttpClientErrorException.create( HttpStatus.BAD_REQUEST, "'name' must not be null.", null, null, StandardCharsets.UTF_8 ); if ( StringUtils.isNullOrEmpty( addressInfo.getAddress() ) ) throw HttpClientErrorException.create( HttpStatus.BAD_REQUEST, "'address' must not be null.", null, null, StandardCharsets.UTF_8 ); if ( addressInfo.getMobile() == null ) throw HttpClientErrorException.create( HttpStatus.BAD_REQUEST, "'mobile' must not be null.", null, null, StandardCharsets.UTF_8 ); } } 新增地址 ################################## # 新增地址 ################################## POST http://<host>/<context-path>/address/new @PostMapping(value = URL_NEW_ADDRESS, consumes = MediaType.APPLICATION_JSON_UTF8_VALUE, produces = MediaType.APPLICATION_JSON_UTF8_VALUE) public ResponseEntity<BaseResponseEntity<String>> newAddress( @RequestBody DeliverAddress addressInfo ) { log.info( "Request-Method: POST, Request-Path: /new, addressInfo: {}", addressInfo.toString() ); checkIn( addressInfo ); protobuf2RedisService.newAddress( addressInfo ); BaseResponseEntity.BaseResponseEntityBuilder<String> builder = BaseResponseEntity.builder(); return ResponseEntity.ok() .body( builder.code( Code.OK.getCode() ) .data( "Address added successfully" ) .build() ); }; 查询收货地址 ################################## # 查询收货地址 ################################## GET http://<host>/<context-path>/address/fetch @GetMapping(value = URL_FETCH_ADDRESS, consumes = MediaType.APPLICATION_FORM_URLENCODED_VALUE, produces = MediaType.APPLICATION_JSON_UTF8_VALUE) public ResponseEntity<BaseResponseEntity<?>> fetchAddress( @RequestParam(value = "uid", required = false) String uid ) { log.info( "Request-Method: GET, Request-Path: /fetch, uid: {}", uid ); // check parameter must not be null if ( StringUtils.isNullOrEmpty( uid ) ) { throw HttpClientErrorException.create( HttpStatus.BAD_REQUEST, "'uid' must not be null.", null, null, StandardCharsets.UTF_8 ); } List<DeliverAddress> data = protobuf2RedisService.fetchAddress( uid ); BaseResponseEntity.BaseResponseEntityBuilder<Object> entityBuilder = BaseResponseEntity.builder() .code( Code.OK.getCode() ); if ( data.isEmpty() ) { return ResponseEntity.ok() .body( entityBuilder.data( "No result" ).build() ); } return ResponseEntity.ok() .body( entityBuilder.data( data ).build() ); } 修改收货地址 ################################## # 修改收货地址 ################################## POST http://<host>/<context-path>/address/modify @PostMapping(value = URL_MODIFY_ADDRESS, consumes = MediaType.APPLICATION_JSON_UTF8_VALUE, produces = MediaType.APPLICATION_JSON_UTF8_VALUE) public ResponseEntity<BaseResponseEntity<String>> modifyAddress( @RequestBody DeliverAddress addressInfo ) { log.info( "Request-Method: POST, Request-Path: /modify, addressInfo: {}", addressInfo.toString() ); // check parameter must not be null if ( addressInfo.getId() == null ) throw HttpClientErrorException.create( HttpStatus.BAD_REQUEST, "'id' must not be null.", null, null, StandardCharsets.UTF_8 ); checkIn( addressInfo ); protobuf2RedisService.modifyAddress( addressInfo ); BaseResponseEntity.BaseResponseEntityBuilder<String> builder = BaseResponseEntity.builder(); return ResponseEntity.ok() .body( builder.code( Code.OK.getCode() ) .data( "Address modified successfully" ) .build() ); } 删除收货地址 ################################## # 删除收货地址 ################################## DELETE http://<host>/<context-path>/address/evict/{id} @DeleteMapping(value = URL_EVICT_ADDRESS, consumes = MediaType.APPLICATION_FORM_URLENCODED_VALUE, produces = MediaType.APPLICATION_JSON_UTF8_VALUE) public ResponseEntity<BaseResponseEntity<String>> evictAddress( @PathVariable(value = "id", required = false) Long id, @RequestParam(value = "uid", required = false) String uid ) { log.info( "Request-Method: DELETE, Request-Path: /evict, id: {}, uid: {}", id, uid ); // check parameter must not be null if ( id == null ) throw HttpClientErrorException.create( HttpStatus.BAD_REQUEST, "'id' must not be null.", null, null, StandardCharsets.UTF_8 ); if ( StringUtils.isNullOrEmpty( uid ) ) throw HttpClientErrorException.create( HttpStatus.BAD_REQUEST, "'uid' must not be null.", null, null, StandardCharsets.UTF_8 ); protobuf2RedisService.evictAddress( id, uid ); BaseResponseEntity.BaseResponseEntityBuilder<String> builder = BaseResponseEntity.builder(); return ResponseEntity.ok() .body( builder.code( Code.OK.getCode() ) .data( "Address deleted successfully" ) .build() ); }

DAO层使用Cache注解

package com.yorozuyas.demo.dao.impl; ... import com.yorozuyas.demo.dao.Protobuf2RedisDAO; import com.yorozuyas.demo.model.DeliverAddress; @Component public class Protobuf2RedisDAOImpl implements Protobuf2RedisDAO { /** * name space */ private static final String NEW_ADDRESS = "com.yorozuyas.demo.dao.Protobuf2RedisMapper.newAddress"; private static final String FETCH_ADDRESS = "com.yorozuyas.demo.dao.Protobuf2RedisMapper.fetchAddress"; private static final String MODIFY_ADDRESS = "com.yorozuyas.demo.dao.Protobuf2RedisMapper.modifyAddress"; private static final String EVICT_ADDRESS = "com.yorozuyas.demo.dao.Protobuf2RedisMapper.evictAddress"; @Autowired private SqlSessionTemplate sqlSessionTemplate; @Override @Cacheable(cacheNames = "address", key = "'address:' + #addressInfo.getUid()", unless = "#result.isEmpty()") public List<DeliverAddress> newAddress( DeliverAddress addressInfo ) { sqlSessionTemplate.insert( NEW_ADDRESS, addressInfo ); return sqlSessionTemplate.selectList( FETCH_ADDRESS, addressInfo.getUid() ); } @Override @Cacheable(cacheNames = "address", key = "'address:' + #uid", unless = "#result.isEmpty()") public List<DeliverAddress> fetchAddress( String uid ) { return sqlSessionTemplate.selectList( FETCH_ADDRESS, uid ); } @Override @CachePut(cacheNames = "address", key = "'address:' + #addressInfo.getUid()", unless = "#result.isEmpty()") public List<DeliverAddress> modifyAddress( DeliverAddress addressInfo ) { sqlSessionTemplate.update( MODIFY_ADDRESS, addressInfo ); return sqlSessionTemplate.selectList( FETCH_ADDRESS, addressInfo.getUid() ); } @Override @CacheEvict(cacheNames = "address", key = "'address:' + #uid") public int evictAddress( Long id, String uid ) { final Map<String, Object> map = new HashMap<String, Object>( 2 ); map.put( "id", id ); map.put( "uid", uid ); return sqlSessionTemplate.delete( EVICT_ADDRESS, map ); } }

测试

运行redis，执行redis-cli进入redis并执行monitor监视连接请求。

运行MySQL，并执行初始化sql。

运行SpringBoot。

调用查询收货地址api

查看控制台输出信息，输出如下：

2020-07-04 19:43:45.752 14548 --- [ioEventLoop-4-1] DEBUG io.lettuce.core.RedisClient - Connecting to Redis at www.yorozuyas.com:6379: Success 2020-07-04 19:43:45.754 14548 --- [ioEventLoop-4-1] DEBUG io.lettuce.core.RedisChannelHandler - dispatching command AsyncCommand [type=AUTH, output=StatusOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.AsyncCommand] 2020-07-04 19:43:45.755 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.DefaultEndpoint - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, epid=0x1] write() writeAndFlush command AsyncCommand [type=AUTH, output=StatusOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.AsyncCommand] 2020-07-04 19:43:45.756 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] write(ctx, AsyncCommand [type=AUTH, output=StatusOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.AsyncCommand], promise) 2020-07-04 19:43:45.759 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandEncoder - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379] writing command AsyncCommand [type=AUTH, output=StatusOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.AsyncCommand] 2020-07-04 19:43:45.759 14548 --- [ioEventLoop-4-1] TRACE i.l.core.protocol.CommandEncoder - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379] Sent: *2 $4 AUTH $6 password 2020-07-04 19:43:45.762 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.DefaultEndpoint - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, epid=0x1] write() done 2020-07-04 19:43:45.762 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.ConnectionWatchdog - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, last known addr=www.yorozuyas.com/121.36.211.56:6379] userEventTriggered(ctx, io.lettuce.core.ConnectionEvents$Activated@8ff7c65) 2020-07-04 19:43:45.762 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.ConnectionWatchdog - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, last known addr=www.yorozuyas.com/121.36.211.56:6379] userEventTriggered(ctx, io.lettuce.core.ConnectionEvents$Activated@8ff7c65) 2020-07-04 19:43:45.769 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Received: 5 bytes, 1 commands in the stack 2020-07-04 19:43:45.769 14548 --- [ioEventLoop-4-1] TRACE i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Buffer: +OK 2020-07-04 19:43:45.769 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Stack contains: 1 commands 2020-07-04 19:43:45.769 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.RedisStateMachine - Decode AsyncCommand [type=AUTH, output=StatusOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.AsyncCommand] 2020-07-04 19:43:45.772 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.RedisStateMachine - Decoded AsyncCommand [type=AUTH, output=StatusOutput [output=OK, error='null'], commandType=io.lettuce.core.protocol.AsyncCommand], empty stack: true 2020-07-04 19:43:45.774 14548 --- [nio-8080-exec-2] DEBUG io.lettuce.core.RedisChannelHandler - dispatching command AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 19:43:45.774 14548 --- [nio-8080-exec-2] DEBUG i.l.core.protocol.DefaultEndpoint - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, epid=0x1] write() writeAndFlush command AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 19:43:45.775 14548 --- [nio-8080-exec-2] DEBUG i.l.core.protocol.DefaultEndpoint - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, epid=0x1] write() done 2020-07-04 19:43:45.775 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] write(ctx, AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command], promise) 2020-07-04 19:43:45.776 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandEncoder - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379] writing command AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 19:43:45.776 14548 --- [ioEventLoop-4-1] TRACE i.l.core.protocol.CommandEncoder - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379] Sent: *2 $3 GET $16 address:zhangsan 2020-07-04 19:43:45.786 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Received: 5 bytes, 1 commands in the stack 2020-07-04 19:43:45.786 14548 --- [ioEventLoop-4-1] TRACE i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Buffer: $-1 2020-07-04 19:43:45.786 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Stack contains: 1 commands 2020-07-04 19:43:45.786 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.RedisStateMachine - Decode AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 19:43:45.786 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.RedisStateMachine - Decoded AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command], empty stack: true 2020-07-04 19:43:45.817 14548 --- [nio-8080-exec-2] DEBUG c.y.d.d.P.fetchAddress - ==> Preparing: SELECT id, uid, name, address, mobile FROM tb_deliver_address WHERE uid = ? 2020-07-04 19:43:45.834 14548 --- [nio-8080-exec-2] DEBUG c.y.d.d.P.fetchAddress - ==> Parameters: zhangsan(String) 2020-07-04 19:43:45.857 14548 --- [nio-8080-exec-2] DEBUG c.y.d.d.P.fetchAddress - <== Total: 4 2020-07-04 19:43:45.875 14548 --- [nio-8080-exec-2] DEBUG io.lettuce.core.RedisChannelHandler - dispatching command AsyncCommand [type=SET, output=StatusOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 19:43:45.876 14548 --- [nio-8080-exec-2] DEBUG i.l.core.protocol.DefaultEndpoint - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, epid=0x1] write() writeAndFlush command AsyncCommand [type=SET, output=StatusOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 19:43:45.876 14548 --- [nio-8080-exec-2] DEBUG i.l.core.protocol.DefaultEndpoint - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, epid=0x1] write() done 2020-07-04 19:43:45.876 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] write(ctx, AsyncCommand [type=SET, output=StatusOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command], promise) 2020-07-04 19:43:45.877 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandEncoder - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379] writing command AsyncCommand [type=SET, output=StatusOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 19:43:45.877 14548 --- [ioEventLoop-4-1] TRACE i.l.core.protocol.CommandEncoder - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379] Sent: *5 $3 SET $16 address:zhangsan $538 ? ArrayList?'com.yorozuyas.demo.model.DeliverAddresszhangsan鎴戠殑濮愬"-婀栧寳鐪佹姹夊競姹熸眽鍖哄彂灞曞ぇ閬?185鍙?(靖瘙-?'com.yorozuyas.demo.model.DeliverAddresszhangsan鎴戠殑濡瑰"6瀹夊窘鐪佸悎鑲ュ競缁忔祹鎶?鏈紑鍙戝尯鑺欒搲璺?678鍙?(扛瘙-?'com.yorozuyas.demo.model.DeliverAddresszhangsan鎴戠殑寮熷紵"9娴欐睙鐪佹澀宸炲競婊ㄦ睙鍖烘睙鍗楀ぇ閬撻緳婀栨花姹熷ぉ琛?(栏瘙-?'com.yorozuyas.demo.model.DeliverAddresszhangsan寮犱笁">骞夸笢鐪佷經灞卞競椤哄痉鍖哄ぇ鑹繋瀹捐矾纰ф鍥烽捇鐭虫咕(粮瘙- $2 PX $7 1800000

从控制台可以看出，在请求过来之后，SpringCache注解起作用了，首先是执行登录请求，然后查询address:zhangsan，结果返回 Buffer: $-1，说明缓存中没有这条信息，那么就执行数据库查询操作，最后把查询结果添加到缓存中。

查看redis监控 可以看到先有一个登录请求过来，然后执行了GET查询操作，最后执行了SET添加操作，和上面分析一致。

再调用一次查询收货地址api

查看控制台输出

2020-07-04 20:13:36.682 14548 --- [nio-8080-exec-5] INFO c.y.d.c.Protobuf2RedisController - Request-Method: GET, Request-Path: /fetch, uid: zhangsan 2020-07-04 20:13:36.705 14548 --- [nio-8080-exec-5] DEBUG io.lettuce.core.RedisChannelHandler - dispatching command AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 20:13:36.705 14548 --- [nio-8080-exec-5] DEBUG i.l.core.protocol.DefaultEndpoint - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, epid=0x1] write() writeAndFlush command AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 20:13:36.706 14548 --- [nio-8080-exec-5] DEBUG i.l.core.protocol.DefaultEndpoint - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, epid=0x1] write() done 2020-07-04 20:13:36.706 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] write(ctx, AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command], promise) 2020-07-04 20:13:36.706 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandEncoder - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379] writing command AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 20:13:36.707 14548 --- [ioEventLoop-4-1] TRACE i.l.core.protocol.CommandEncoder - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379] Sent: *2 $3 GET $16 address:zhangsan 2020-07-04 20:13:36.715 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Received: 512 bytes, 1 commands in the stack 2020-07-04 20:13:36.715 14548 --- [ioEventLoop-4-1] TRACE i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Buffer: $538 ? ArrayList?'com.yorozuyas.demo.model.DeliverAddresszhangsan鎴戠殑濮愬"-婀栧寳鐪佹姹夊競姹熸眽鍖哄彂灞曞ぇ閬?185鍙?(靖瘙-?'com.yorozuyas.demo.model.DeliverAddresszhangsan鎴戠殑濡瑰"6瀹夊窘鐪佸悎鑲ュ競缁忔祹鎶?鏈紑鍙戝尯鑺欒搲璺?678鍙?(扛瘙-?'com.yorozuyas.demo.model.DeliverAddresszhangsan鎴戠殑寮熷紵"9娴欐睙鐪佹澀宸炲競婊ㄦ睙鍖烘睙鍗楀ぇ閬撻緳婀栨花姹熷ぉ琛?(栏瘙-?'com.yorozuyas.demo.model.DeliverAddresszhangsan寮犱笁">骞夸笢鐪佷經灞卞競椤哄痉鍖哄ぇ鑹繋瀹 2020-07-04 20:13:36.715 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Stack contains: 1 commands 2020-07-04 20:13:36.715 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.RedisStateMachine - Decode AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 20:13:36.715 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.RedisStateMachine - Decoded AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command], empty stack: false 2020-07-04 20:13:36.716 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Received: 34 bytes, 1 commands in the stack 2020-07-04 20:13:36.716 14548 --- [ioEventLoop-4-1] TRACE i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Buffer: 捐矾纰ф鍥烽捇鐭虫咕(粮瘙- 2020-07-04 20:13:36.716 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.CommandHandler - [channel=0xc4d24ef3, /192.168.3.16:50627 -> www.yorozuyas.com/121.36.211.56:6379, chid=0x1] Stack contains: 1 commands 2020-07-04 20:13:36.716 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.RedisStateMachine - Decode AsyncCommand [type=GET, output=ValueOutput [output=null, error='null'], commandType=io.lettuce.core.protocol.Command] 2020-07-04 20:13:36.716 14548 --- [ioEventLoop-4-1] DEBUG i.l.core.protocol.RedisStateMachine - Decoded AsyncCommand [type=GET, output=ValueOutput [output=[B@2e463c2d, error='null'], commandType=io.lettuce.core.protocol.Command], empty stack: true

可以看到，执行了查询Redis，并得到结果Buffer: $538，然后就返回数据，没有执行数据库操作（没有打印SQL信息判断得知）。 另外，redis缓存的数据分别打印了两次，出现了Netty的读写半包，有兴趣的同学可以学习一下Netty，可参考《Netty权威指南》或《Netty实战》。

查看redis监控 监控得知，执行了查询操作，后面就没有其他输出信息了，证明查到数据了。其他接口就不再测试了，可自行进行测试。

结束语

总体来说Spring对Redis的集成度很高，在没有特殊需求的情况下，引入类库就可直接使用Spring注解进行缓存集成。而其他缓存如Memcached就没有得到Spring的这般对待了，要想集成Spring注解，需要自己实现CacheManager和Cache接口。文中提供的org.springframework.data.redis.cache.RedisCacheConfiguration，针对的是全局的，也就是默认的。redis可以针对不同的命名空间提供不同的配置，具体需要自定义一个RedisCacheManager Bean，并执行RedisCacheManagerBuilder方法withInitialCacheConfigurations添加Map<String , RedisCacheConfiguration>。