springboot2.0 redis EnableCaching的配置和使用

時間 2019-11-06

標籤 springboot2.0 springboot redis enablecaching 配置使用欄目 Spring 简体版

原文原文鏈接

1、前言

　　關於EnableCaching最簡單使用，我的感受只需提供一個CacheManager的一個實例就行了。springboot爲咱們提供了cache相關的自動配置。引入cache模塊，以下。html

2、maven依賴

<dependency>

    <groupId>org.springframework.boot</groupId>

    <artifactId>spring-boot-starter-cache</artifactId>

</dependency>

<dependency>

    <groupId>org.springframework.boot</groupId>

    <artifactId>spring-boot-starter-data-redis</artifactId>

</dependency>

<dependency>

    <groupId>org.springframework.integration</groupId>

    <artifactId>spring-integration-redis</artifactId>

</dependency>

<dependency>

    <groupId>redis.clients</groupId>

    <artifactId>jedis</artifactId>

</dependency>

3、緩存類型

　　本人也僅僅使用了redis、guava、ehcache。更多詳情請參考 spring cache官方文檔。java

https://docs.spring.io/spring-boot/docs/current/reference/html/boot-features-caching.html

4、經常使用註解

　　@Cacheable 觸發緩存填充redis

　　@CacheEvict 觸發緩存驅逐spring

　　@CachePut 更新緩存而不會干擾方法執行數組

　　@Caching 從新組合要在方法上應用的多個緩存操做緩存

　　@CacheConfig 在類級別共享一些常見的緩存相關設置安全

5、Spring Cache提供的SpEL上下文數據

　　下表直接摘自Spring官方文檔：springboot

名字	位置	描述	示例
methodName	root對象	當前被調用的方法名	#root.methodName
method	root對象	當前被調用的方法	#root.method.name
target	root對象	當前被調用的目標對象	#root.target
targetClass	root對象	當前被調用的目標對象類	#root.targetClass
args	root對象	當前被調用的方法的參數列表	#root.args[0]
caches	root對象	當前方法調用使用的緩存列表（如@Cacheable(value={"cache1", "cache2"})），則有兩個cache	#root.caches[0].name
argument name	執行上下文	當前被調用的方法的參數，如findById(Long id)，咱們能夠經過#id拿到參數	#user.id
result	執行上下文	方法執行後的返回值（僅當方法執行以後的判斷有效，如‘unless’，'cache evict'的beforeInvocation=false）	#result

6、RedisCacheManager配置

　　基於jedis

@Configuration
@EnableCaching
public class RedisCacheConfig extends CachingConfigurerSupport {
    @Autowired
    private RedisProperties redisProperties;
    @Bean
    public JedisConnectionFactory jedisConnectionFactory() {
        // 獲取服務器數組(這裏要相信本身的輸入，因此沒有考慮空指針問題)
        String[] serverArray = redisProperties.getClusterNodes().split(",");
        RedisClusterConfiguration redisClusterConfiguration = new 
            RedisClusterConfiguration(Arrays.asList(serverArray));
        JedisPoolConfig jedisPoolConfig = new JedisPoolConfig();
        // 最大空閒鏈接數, 默認8個
        jedisPoolConfig.setMaxIdle(100);
        // 最大鏈接數, 默認8個
        jedisPoolConfig.setMaxTotal(500);
        // 最小空閒鏈接數, 默認0
        jedisPoolConfig.setMinIdle(0);
        // 獲取鏈接時的最大等待毫秒數(若是設置爲阻塞時BlockWhenExhausted),若是超時就拋異常, 小於零:阻塞不肯定的時間,
        // 默認-1
        jedisPoolConfig.setMaxWaitMillis(2000); // 設置2秒
        // 對拿到的connection進行validateObject校驗
        jedisPoolConfig.setTestOnBorrow(true);
        return new JedisConnectionFactory(redisClusterConfiguration
            ,jedisPoolConfig);
    }
    /**     * 注入redis template     *     * @return     */
    @Bean
    @Qualifier("redisTemplate")
    public RedisTemplate redisTemplate(
        JedisConnectionFactoryjedisConnectionFactory
        , Jackson2JsonRedisSerializer jackson2JsonRedisSerializer) {
        RedisTemplate template = new RedisTemplate();
        template.setConnectionFactory(jedisConnectionFactory);
        template.setKeySerializer(new JdkSerializationRedisSerializer());
        template.setValueSerializer(jackson2JsonRedisSerializer);
        template.afterPropertiesSet();
        return template;
    }
    /**     * redis cache manager     *     * @return     */
    @Bean
    @Primary
    public RedisCacheManager redisCacheManager(
        JedisConnectionFactory jedisConnectionFactory
        , ObjectProvider<List<RedisCacheConfigurationProvider>> 
            configurationProvider) {
        Map<String, RedisCacheConfiguration> redisCacheConfigurationMap = 
            Maps.newHashMap();
        List<RedisCacheConfigurationProvider> configurations
             = configurationProvider.getIfAvailable();
        if (!CollectionUtils.isEmpty(configurations)) {
            for (RedisCacheConfigurationProvider configuration : configurations) {
                redisCacheConfigurationMap.putAll(configuration.resolve());
            }
        }
        RedisCacheManager cacheManager = RedisCacheManager.
            RedisCacheManagerBuilder.fromConnectionFactory(jedisConnectionFactory)
                .cacheDefaults(resovleRedisCacheConfiguration(Duration.
                    ofSeconds(300), JacksonHelper.genJavaType(Object.class)))
                .withInitialCacheConfigurations(redisCacheConfigurationMap)
                .build();
        return cacheManager;
    }
    private static RedisCacheConfiguration resovleRedisCacheConfiguration(Duration duration, JavaType javaType) {
        return RedisCacheConfiguration.defaultCacheConfig()
                .serializeKeysWith(RedisSerializationContext
                    .SerializationPair
                    .fromSerializer(new StringRedisSerializer()))
                .serializeValuesWith(RedisSerializationContext
                    .SerializationPair.fromSerializer(
                        new Jackson2JsonRedisSerializer<>(javaType)))
                .entryTtl(duration);
    }
    /**     * 配置一個序列器, 將對象序列化爲字符串存儲, 和將對象反序列化爲對象     */
    @Bean
    public Jackson2JsonRedisSerializer jackson2JsonRedisSerializer() {
        Jackson2JsonRedisSerializer jackson2JsonRedisSerializer = new Jackson2JsonRedisSerializer(Object.class);
        ObjectMapper om = new ObjectMapper();
        om.setVisibility(PropertyAccessor.ALL, JsonAutoDetect.Visibility.ANY);
        om.enableDefaultTyping(ObjectMapper.DefaultTyping.NON_FINAL);
        jackson2JsonRedisSerializer.setObjectMapper(om);
        return jackson2JsonRedisSerializer;
    }
    public static abstract class RedisCacheConfigurationProvider {
        // key = 緩存名稱， value = 緩存時間 和 緩存類型
        protected Map<String, Pair<Duration, JavaType>> configs;
        protected abstract void initConfigs();
        public Map<String, RedisCacheConfiguration> resolve() {
            initConfigs();
            Assert.notEmpty(configs, "RedisCacheConfigurationProvider 配置不能爲空...");
            Map<String, RedisCacheConfiguration> result = Maps.newHashMap();
            configs.forEach((cacheName, pair) -> result.put(cacheName, 
                resovleRedisCacheConfiguration(pair.getKey(), pair.getValue())));
            return result;
        }
    }

}

　　基於Lettuce

@Configuration
@EnableCaching
public class RedisCacheConfig extends CachingConfigurerSupport {

　　@Autowired
    private RedisProperties redisProperties;
    @Bean
    public LettuceConnectionFactory lettuceConnectionFactory() {
        String[] serverArray = redisProperties.getClusterNodes().split(",");// 獲取服務器數組(這裏要相信本身的輸入，因此沒有考慮空指針問題)
        RedisClusterConfiguration redisClusterConfiguration = new RedisClusterConfiguration(Arrays.asList(serverArray));
        GenericObjectPoolConfig poolConfig = new GenericObjectPoolConfig();
        // 最大空閒鏈接數, 默認8個
        poolConfig.setMaxIdle(100);
        // 最大鏈接數, 默認8個
        poolConfig.setMaxTotal(500);
        // 最小空閒鏈接數, 默認0
        poolConfig.setMinIdle(0);
        LettuceClientConfiguration lettuceClientConfiguration = LettucePoolingClientConfiguration.builder()
                .commandTimeout(Duration.ofSeconds(15))
                .poolConfig(poolConfig)
                .shutdownTimeout(Duration.ZERO)
                .build();
        return new LettuceConnectionFactory(redisClusterConfiguration, lettuceClientConfiguration);
    }
    /**     * 注入redis template     *     * @return     */
    @Bean
    @Qualifier("redisTemplate")
    public RedisTemplate redisTemplate(LettuceConnectionFactory lettuceConnectionFactory, Jackson2JsonRedisSerializer jackson2JsonRedisSerializer) {
        RedisTemplate template = new RedisTemplate();
        template.setConnectionFactory(lettuceConnectionFactory);
        template.setKeySerializer(new JdkSerializationRedisSerializer());
        template.setValueSerializer(jackson2JsonRedisSerializer);
        template.afterPropertiesSet();
        return template;
    }
    /**     * redis cache manager     *     * @return     */
    @Bean
    @Primary
    public RedisCacheManager redisCacheManager(LettuceConnectionFactory lettuceConnectionFactory, ObjectProvider<List<RedisCacheConfigurationProvider>> configurationProvider) {
        Map<String, RedisCacheConfiguration> redisCacheConfigurationMap = Maps.newHashMap();
        List<RedisCacheConfigurationProvider> configurations = configurationProvider.getIfAvailable();
        if (!CollectionUtils.isEmpty(configurations)) {
            for (RedisCacheConfigurationProvider configuration : configurations) {
                redisCacheConfigurationMap.putAll(configuration.resolve());
            }
        }
        RedisCacheManager cacheManager = RedisCacheManager.RedisCacheManagerBuilder.fromConnectionFactory(lettuceConnectionFactory)
                .cacheDefaults(resovleRedisCacheConfiguration(Duration.ofSeconds(300), JacksonHelper.genJavaType(Object.class)))
                .withInitialCacheConfigurations(redisCacheConfigurationMap)
                .build();
        return cacheManager;
    }
    private static RedisCacheConfiguration resovleRedisCacheConfiguration(Duration duration, JavaType javaType) {
        return RedisCacheConfiguration.defaultCacheConfig()
                .serializeKeysWith(RedisSerializationContext.SerializationPair.fromSerializer(new StringRedisSerializer()))
                .serializeValuesWith(RedisSerializationContext.SerializationPair.fromSerializer(new Jackson2JsonRedisSerializer<>(javaType)))
                .entryTtl(duration);
    }
    @Bean
    public RedisLockRegistry redisLockRegistry(LettuceConnectionFactory lettuceConnectionFactory) {
        return new RedisLockRegistry(lettuceConnectionFactory, "recharge-plateform", 60000 * 20);
    }
    /**     * 配置一個序列器, 將對象序列化爲字符串存儲, 和將對象反序列化爲對象     */
    @Bean
    public Jackson2JsonRedisSerializer jackson2JsonRedisSerializer() {
        Jackson2JsonRedisSerializer jackson2JsonRedisSerializer = new Jackson2JsonRedisSerializer(Object.class);
        ObjectMapper om = new ObjectMapper();
        om.setVisibility(PropertyAccessor.ALL, JsonAutoDetect.Visibility.ANY);
        om.enableDefaultTyping(ObjectMapper.DefaultTyping.NON_FINAL);
        jackson2JsonRedisSerializer.setObjectMapper(om);
        return jackson2JsonRedisSerializer;
    }
    public static abstract class RedisCacheConfigurationProvider {
        // key = 緩存名稱， value = 緩存時間 和 緩存類型
        protected Map<String, Pair<Duration, JavaType>> configs;
        protected abstract void initConfigs();
        public Map<String, RedisCacheConfiguration> resolve() {
            initConfigs();
            Assert.notEmpty(configs, "RedisCacheConfigurationProvider 配置不能爲空...");
            Map<String, RedisCacheConfiguration> result = Maps.newHashMap();
            configs.forEach((cacheName, pair) -> result.put(cacheName, resovleRedisCacheConfiguration(pair.getKey(), pair.getValue())));
            return result;
        }
    }

}

　　Jedis和Lettuce比較

　　Jedis 是直連模式，在多個線程間共享一個 Jedis 實例時是線程不安全的，若是想要在多線程環境下使用 Jedis，須要使用鏈接池，服務器

　　每一個線程都去拿本身的 Jedis 實例，當鏈接數量增多時，物理鏈接成本就較高了。網絡

　　Lettuce的鏈接是基於Netty的，鏈接實例能夠在多個線程間共享，

　　因此，一個多線程的應用可使用同一個鏈接實例，而不用擔憂併發線程的數量。固然這個也是可伸縮的設計，一個鏈接實例不夠的狀況也能夠按需增長鏈接實例。經過異步的方式可讓咱們更好的利用系統資源，而不用浪費線程等待網絡或磁盤I/O。

　　只在基於Lettuce的配置中，加入了RedisLockRegistry對應bean的配置，因爲在集羣的模式下，基於Jedis的配置下，經過RedisLockRegistry 獲取分佈式鎖的時候報錯：

EvalSha is not supported in cluster environment

　　具體的解決方案就是切換至基於Lettuce的配置，請參考

https://stackoverflow.com/questions/47092475/spring-boot-redistemplate-execute-sc

　　RedisCacheConfigurationProvider 做用爲不一樣的cache提供特定的緩存時間以及key、value序列化和反序列化的方式。具體使用方式以下。

@Component

public class CouponRedisCacheConfigurationProvider extends RedisCacheConfig.RedisCacheConfigurationProvider {

@Override
    protected void initConfigs() {
        this.configs = Maps.newHashMap();
        this.configs.put(CouponConstants.COUPON_ALL_CACHE, new Pair<>(Duration.ofHours(1), JacksonHelper.genMapType(HashMap.class, Integer.class, Coupon.class)));
        this.configs.put(CouponConstants.COUPON_GOOD_CACHE, new Pair<>(Duration.ofHours(1), JacksonHelper.genCollectionType(List.class, String.class)));

        this.configs.put(CouponConstants.COUPON_HANDLE_TELEPHONE_STATUS_CACHE, new Pair<>(Duration.ofMinutes(10), JacksonHelper.genCollectionType(List.class, CouponHandle.class)));
        this.configs.put(CouponConstants.COUPON_HANDLE_TELEPHONE_GOOD_CACHE, new Pair<>(Duration.ofMinutes(10), JacksonHelper.genCollectionType(List.class, CouponHandle.class)));
    }

}

　　JacksonHelper 做用是根據不一樣的類型的對象獲取對應的JavaType對象，在構造RedisTempalte序列化和反序列化器Jackson2JsonRedisSerializer對象須要。具體代碼以下。

public class JacksonHelper {
    private static Logger LOGGER = LoggerFactory.getLogger(JacksonHelper.class);

    private static final SimpleModule module = initModule();
    private static final ObjectMapper objectMapper;
    private static final ObjectMapper prettyMapper;

    public JacksonHelper() {
    }

    private static SimpleModule initModule() {
        return (new SimpleModule()).addSerializer(BigDecimal.class, new BigDecimalSerializer())
                .addSerializer(LocalTime.class, new LocalTimeSerializer())
                .addDeserializer(LocalTime.class, new LocalTimeDeserializer())
                .addSerializer(LocalDate.class, new LocalDateSerializer())
                .addDeserializer(LocalDate.class, new LocalDateDeserializer())
                .addSerializer(LocalDateTime.class, new LocalDateTimeSerializer())
                .addDeserializer(LocalDateTime.class, new LocalDateTimeDeserializer())
                .addSerializer(Date.class, new DateSerializer())
                .addDeserializer(Date.class, new DateDeserializer());
    }

    public static JavaType genJavaType(TypeReference<?> typeReference) {
        return getObjectMapper().getTypeFactory().constructType(typeReference.getType());
    }

    public static JavaType genJavaType(Class<?> clazz) {
        return getObjectMapper().getTypeFactory().constructType(clazz);
    }

    public static JavaType genCollectionType(Class<? extends Collection> collectionClazz, Class<?> javaClazz) {
        return getObjectMapper().getTypeFactory().constructCollectionType(collectionClazz, javaClazz);
    }

    public static JavaType genMapType(Class<? extends Map> mapClazz, Class<?> keyClass, Class<?> valueClazz) {
        return getObjectMapper().getTypeFactory().constructMapType(mapClazz, keyClass, valueClazz);
    }

    public static ObjectMapper getObjectMapper() {
        return objectMapper;
    }

    public static ObjectMapper getPrettyMapper() {
        return prettyMapper;
    }

    static {
        objectMapper = (new ObjectMapper()).registerModule(module).configure(DeserializationFeature.FAIL_ON_UNKNOWN_PROPERTIES, false).configure(DeserializationFeature.READ_UNKNOWN_ENUM_VALUES_AS_NULL, true).configure(SerializationFeature.FAIL_ON_EMPTY_BEANS, false);
        prettyMapper = objectMapper.copy().configure(SerializationFeature.INDENT_OUTPUT, true);
    }

}



class LocalDateDeserializer extends JsonDeserializer<LocalDate> {

    public LocalDateDeserializer() {
    }

    public LocalDate deserialize(JsonParser jp, DeserializationContext ctxt) throws IOException {
        String dateString = ((JsonNode) jp.getCodec().readTree(jp)).asText();
        return LocalDate.parse(dateString, DateTimeFormatter.ISO_LOCAL_DATE);
    }

}



class DateDeserializer extends JsonDeserializer<Date> {

    public DateDeserializer() {
    }

    public Date deserialize(JsonParser jp, DeserializationContext ctxt) throws IOException {
        String dateTimeStr = ((JsonNode) jp.getCodec().readTree(jp)).asText();
        SimpleDateFormat sdf = new SimpleDateFormat(CouponConstants.DATE_TIME_FORMATER);
        ParsePosition pos = new ParsePosition(0);
        return sdf.parse(dateTimeStr, pos);
    }

}



class LocalDateTimeDeserializer extends JsonDeserializer<LocalDateTime> {

    public LocalDateTimeDeserializer() {
    }

    public LocalDateTime deserialize(JsonParser jp, DeserializationContext ctxt) throws IOException {
        String dateTimeStr = ((JsonNode) jp.getCodec().readTree(jp)).asText();
        return LocalDateTime.parse(dateTimeStr, DateTimeFormatter.ISO_LOCAL_DATE_TIME);
    }

}



class LocalTimeDeserializer extends JsonDeserializer<LocalTime> {

    public LocalTimeDeserializer() {
    }

    public LocalTime deserialize(JsonParser jp, DeserializationContext ctxt) throws IOException {
        String dateString = ((JsonNode) jp.getCodec().readTree(jp)).asText();
        return LocalTime.parse(dateString, DateTimeFormatter.ISO_LOCAL_TIME);
    }

}



class BigDecimalSerializer extends JsonSerializer<BigDecimal> {

    public BigDecimalSerializer() {
    }

    public void serialize(BigDecimal value, JsonGenerator jgen, SerializerProvider provider) throws IOException {
        jgen.writeString(value.toString());
    }

}



class LocalDateSerializer extends JsonSerializer<LocalDate> {

    public LocalDateSerializer() {
    }

    public void serialize(LocalDate value, JsonGenerator jgen, SerializerProvider provider) throws IOException {
        jgen.writeString(DateTimeFormatter.ISO_LOCAL_DATE.format(value));
    }

}



class LocalDateTimeSerializer extends JsonSerializer<LocalDateTime> {

    public LocalDateTimeSerializer() {
    }

    public void serialize(LocalDateTime value, JsonGenerator jgen, SerializerProvider provider) throws IOException {
        jgen.writeString(DateTimeFormatter.ISO_LOCAL_DATE_TIME.format(value));
    }

}



class DateSerializer extends JsonSerializer<Date> {

    public DateSerializer() {
    }

    public void serialize(Date value, JsonGenerator jgen, SerializerProvider provider) throws IOException {
        SimpleDateFormat sdf = new SimpleDateFormat(CouponConstants.DATE_TIME_FORMATER);
        jgen.writeString(sdf.format(value));
    }

}



class LocalTimeSerializer extends JsonSerializer<LocalTime> {

    public LocalTimeSerializer() {
    }

    public void serialize(LocalTime value, JsonGenerator jgen, SerializerProvider provider) throws IOException {
        jgen.writeString(DateTimeFormatter.ISO_LOCAL_TIME.format(value));
    }

}

　　業務代碼

@Component
public class ServiceImpl {  

　　 @Override
    @CacheEvict(cacheNames = CouponConstants.COUPON_HANDLE_TELEPHONE_STATUS_CACHE, key = "#telephone+'#'+#status", beforeInvocation = true)
    public void evictCouponHandles(String telephone, Integer status) {

    }

    @Override
    @Cacheable(cacheNames = CouponConstants.COUPON_HANDLE_TELEPHONE_STATUS_CACHE, key = "#telephone+'#'+#status", sync = true)
    public List<CouponHandle> searchCouponHandles(String telephone, Integer status) {
    }

}

　　不一樣的緩存對應不一樣的存儲類型，不一樣的存儲類型對應着不一樣的序列化和反序列化器，這就保證了再調用注有@Cacheable註解的代碼時獲取到的對象不會發生類型轉換錯誤。關於設置不一樣的cache下過時時間以及序列化和反序列器，請參考下面更直接明瞭的例子。

@Configuration

public class RedisCacheConfig {

@Bean
    public KeyGenerator simpleKeyGenerator() {
        return (o, method, objects) -> {
            StringBuilder stringBuilder = new StringBuilder();
            stringBuilder.append(o.getClass().getSimpleName());
            stringBuilder.append(".");
            stringBuilder.append(method.getName());
            stringBuilder.append("[");
            for (Object obj : objects) {
                stringBuilder.append(obj.toString());
            }
            stringBuilder.append("]");

            return stringBuilder.toString();
        };
    }

    @Bean
    public CacheManager cacheManager(RedisConnectionFactory redisConnectionFactory) {
        return new RedisCacheManager(
            RedisCacheWriter.nonLockingRedisCacheWriter(redisConnectionFactory),
            this.getRedisCacheConfigurationWithTtl(600), // 默認策略，未配置的 cache 會使用這個
            this.getRedisCacheConfigurationMap() // 指定 cache 策略
        );
    }

    private Map<String, RedisCacheConfiguration> getRedisCacheConfigurationMap() {
        Map<String, RedisCacheConfiguration> redisCacheConfigurationMap = new HashMap<>();
        redisCacheConfigurationMap.put("UserInfoList", this.getRedisCacheConfigurationWithTtl(3000));
        redisCacheConfigurationMap.put("UserInfoListAnother", this.getRedisCacheConfigurationWithTtl(18000));

        return redisCacheConfigurationMap;
    }

    private RedisCacheConfiguration getRedisCacheConfigurationWithTtl(Integer seconds) {
        Jackson2JsonRedisSerializer<Object> jackson2JsonRedisSerializer = new Jackson2JsonRedisSerializer<>(Object.class);
        ObjectMapper om = new ObjectMapper();
        om.setVisibility(PropertyAccessor.ALL, JsonAutoDetect.Visibility.ANY);
        om.enableDefaultTyping(ObjectMapper.DefaultTyping.NON_FINAL);
        jackson2JsonRedisSerializer.setObjectMapper(om);

        RedisCacheConfiguration redisCacheConfiguration = RedisCacheConfiguration.defaultCacheConfig();
        redisCacheConfiguration = redisCacheConfiguration.serializeValuesWith(
            RedisSerializationContext
                .SerializationPair
                .fromSerializer(jackson2JsonRedisSerializer)
        ).entryTtl(Duration.ofSeconds(seconds));

        return redisCacheConfiguration;
    }

}

@Cacheable(value = "UserInfoList", keyGenerator = "simpleKeyGenerator") // 3000秒
@Cacheable(value = "UserInfoListAnother", keyGenerator = "simpleKeyGenerator") // 18000秒
@Cacheable(value = "DefaultKeyTest", keyGenerator = "simpleKeyGenerator") // 600秒，未指定的cache，使用默認策略

相關標籤/搜索

springboot2.0+redis+fastjson

每日一句

每一个你不满意的现在，都有一个你没有努力的曾经。