Redis 抽奖大转盘的实战示例

时间：2022-04-10 10:33:00|栏目：Redis|点击：次

1. 项目介绍

这是一个基于Spring boot + Mybatis Plus + Redis 的简单案例。

主要是将活动内容、奖品信息、记录信息等缓存到Redis中，然后所有的抽奖过程全部从Redis中做数据的操作。

大致内容很简单，具体操作下面慢慢分析。

2. 项目演示

话不多说，首先上图看看项目效果，如果觉得还行的话咱们就来看看他具体是怎么实现的。

3. 表结构

该项目包含以下四张表，分别是活动表、奖项表、奖品表以及中奖记录表。具体的SQL会在文末给出。

4. 项目搭建

咱们首先先搭建一个标准的Spring boot 项目，直接IDEA创建，然后选择一些相关的依赖即可。

4.1 依赖

该项目主要用到了：Redis，thymeleaf，mybatis-plus等依赖。

<dependencies>

    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-data-redis</artifactId>
    </dependency>

    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-thymeleaf</artifactId>
    </dependency>

    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-web</artifactId>
    </dependency>

    <dependency>
        <groupId>mysql</groupId>
        <artifactId>mysql-connector-java</artifactId>
        <scope>runtime</scope>
    </dependency>

    <dependency>
        <groupId>org.springframework.boot</groupId>
        <artifactId>spring-boot-starter-test</artifactId>
        <scope>test</scope>
    </dependency>

    <dependency>
        <groupId>com.baomidou</groupId>
        <artifactId>mybatis-plus-boot-starter</artifactId>
        <version>3.4.3</version>
    </dependency>

    <dependency>
        <groupId>com.baomidou</groupId>
        <artifactId>mybatis-plus-generator</artifactId>
        <version>3.4.1</version>
    </dependency>

    <dependency>
        <groupId>com.alibaba</groupId>
        <artifactId>fastjson</artifactId>
        <version>1.2.72</version>
    </dependency>

    <dependency>
        <groupId>com.alibaba</groupId>
        <artifactId>druid-spring-boot-starter</artifactId>
        <version>1.1.22</version>
    </dependency>

    <dependency>
        <groupId>org.apache.commons</groupId>
        <artifactId>commons-lang3</artifactId>
        <version>3.9</version>
    </dependency>

    <dependency>
        <groupId>org.projectlombok</groupId>
        <artifactId>lombok</artifactId>
        <version>1.18.12</version>
    </dependency>

    <dependency>
        <groupId>org.apache.commons</groupId>
        <artifactId>commons-pool2</artifactId>
        <version>2.8.0</version>
    </dependency>

    <dependency>
        <groupId>org.mapstruct</groupId>
        <artifactId>mapstruct</artifactId>
        <version>1.4.2.Final</version>
    </dependency>

    <dependency>
        <groupId>org.mapstruct</groupId>
        <artifactId>mapstruct-jdk8</artifactId>
        <version>1.4.2.Final</version>
    </dependency>

    <dependency>
        <groupId>org.mapstruct</groupId>
        <artifactId>mapstruct-processor</artifactId>
        <version>1.4.2.Final</version>
    </dependency>

    <dependency>
        <groupId>joda-time</groupId>
        <artifactId>joda-time</artifactId>
        <version>2.10.6</version>
    </dependency>
</dependencies>

4.2 YML配置

依赖引入之后，我们需要进行相应的配置：数据库连接信息、Redis、mybatis-plus、线程池等。

server:
  port: 8080
  servlet:
    context-path: /
spring:
  datasource:
    druid:
      url: jdbc:mysql://127.0.0.1:3306/test?useUnicode=true&characterEncoding=utf-8&useSSL=false
      username: root
      password: 123456
      driver-class-name: com.mysql.cj.jdbc.Driver
      initial-size: 30
      max-active: 100
      min-idle: 10
      max-wait: 60000
      time-between-eviction-runs-millis: 60000
      min-evictable-idle-time-millis: 300000
      validation-query: SELECT 1 FROM DUAL
      test-while-idle: true
      test-on-borrow: false
      test-on-return: false
      filters: stat,wall
  redis:
    port: 6379
    host: 127.0.0.1
    lettuce:
      pool:
        max-active: -1
        max-idle: 2000
        max-wait: -1
        min-idle: 1
        time-between-eviction-runs: 5000
  mvc:
    view:
      prefix: classpath:/templates/
      suffix: .html
# mybatis-plus
mybatis-plus:
  configuration:
    map-underscore-to-camel-case: true
    auto-mapping-behavior: full
  mapper-locations: classpath*:mapper/**/*Mapper.xml

# 线程池
async:
  executor:
    thread:
      core-pool-size: 6
      max-pool-size: 12
      queue-capacity: 100000
      name-prefix: lottery-service-

4.3 代码生成

这边我们可以直接使用mybatis-plus的代码生成器帮助我们生成一些基础的业务代码，避免这些重复的体力活。

这边贴出相关代码，直接修改数据库连接信息、相关包名模块名即可。

public class MybatisPlusGeneratorConfig {
    public static void main(String[] args) {
        // 代码生成器
        AutoGenerator mpg = new AutoGenerator();

        // 全局配置
        GlobalConfig gc = new GlobalConfig();
        String projectPath = System.getProperty("user.dir");
        gc.setOutputDir(projectPath + "/src/main/java");
        gc.setAuthor("chen");
        gc.setOpen(false);
        //实体属性 Swagger2 注解
        gc.setSwagger2(false);
        mpg.setGlobalConfig(gc);

        // 数据源配置
        DataSourceConfig dsc = new DataSourceConfig();
        dsc.setUrl("jdbc:mysql://127.0.0.1:3306/test?serverTimezone=UTC&useUnicode=true&characterEncoding=utf-8&zeroDateTimeBehavior=convertToNull&useSSL=false&allowPublicKeyRetrieval=true");
        dsc.setDriverName("com.mysql.cj.jdbc.Driver");
        dsc.setUsername("root");
        dsc.setPassword("123456");
        mpg.setDataSource(dsc);

        // 包配置
        PackageConfig pc = new PackageConfig();
//        pc.setModuleName(scanner("模块名"));
        pc.setParent("com.example.lottery");
        pc.setEntity("dal.model");
        pc.setMapper("dal.mapper");
        pc.setService("service");
        pc.setServiceImpl("service.impl");
        mpg.setPackageInfo(pc);


        // 配置模板
        TemplateConfig templateConfig = new TemplateConfig();

        templateConfig.setXml(null);
        mpg.setTemplate(templateConfig);

        // 策略配置
        StrategyConfig strategy = new StrategyConfig();
        strategy.setNaming(NamingStrategy.underline_to_camel);
        strategy.setColumnNaming(NamingStrategy.underline_to_camel);
        strategy.setSuperEntityClass("com.baomidou.mybatisplus.extension.activerecord.Model");
        strategy.setEntityLombokModel(true);
        strategy.setRestControllerStyle(true);

        strategy.setEntityLombokModel(true);
        // 公共父类
//        strategy.setSuperControllerClass("com.baomidou.ant.common.BaseController");
        // 写于父类中的公共字段
//        strategy.setSuperEntityColumns("id");
        strategy.setInclude(scanner("lottery,lottery_item,lottery_prize,lottery_record").split(","));
        strategy.setControllerMappingHyphenStyle(true);
        strategy.setTablePrefix(pc.getModuleName() + "_");
        mpg.setStrategy(strategy);
        mpg.setTemplateEngine(new FreemarkerTemplateEngine());
        mpg.execute();
    }

    public static String scanner(String tip) {
        Scanner scanner = new Scanner(System.in);
        StringBuilder help = new StringBuilder();
        help.append("请输入" + tip + "：");
        System.out.println(help.toString());
        if (scanner.hasNext()) {
            String ipt = scanner.next();
            if (StringUtils.isNotEmpty(ipt)) {
                return ipt;
            }
        }
        throw new MybatisPlusException("请输入正确的" + tip + "！");
    }
}

4.4 Redis 配置

我们如果在代码中使用 RedisTemplate 的话，需要添加相关配置，将其注入到Spring容器中。

@Configuration
public class RedisTemplateConfig {
    @Bean
    public RedisTemplate redisTemplate(RedisConnectionFactory redisConnectionFactory) {
        RedisTemplate<Object, Object> redisTemplate = new RedisTemplate<>();
        redisTemplate.setConnectionFactory(redisConnectionFactory);
        // 使用Jackson2JsonRedisSerialize 替换默认序列化
        Jackson2JsonRedisSerializer jackson2JsonRedisSerializer = new Jackson2JsonRedisSerializer(Object.class);

        ObjectMapper objectMapper = new ObjectMapper();
        objectMapper.setVisibility(PropertyAccessor.ALL, JsonAutoDetect.Visibility.ANY);
        objectMapper.enableDefaultTyping(ObjectMapper.DefaultTyping.NON_FINAL);

        SimpleModule simpleModule = new SimpleModule();
        simpleModule.addSerializer(DateTime.class, new JodaDateTimeJsonSerializer());
        simpleModule.addDeserializer(DateTime.class, new JodaDateTimeJsonDeserializer());
        objectMapper.registerModule(simpleModule);

        jackson2JsonRedisSerializer.setObjectMapper(objectMapper);
        // 设置value的序列化规则和 key的序列化规则
        redisTemplate.setValueSerializer(jackson2JsonRedisSerializer);
        redisTemplate.setKeySerializer(new StringRedisSerializer());

        redisTemplate.setHashKeySerializer(new StringRedisSerializer());
        redisTemplate.setHashValueSerializer(jackson2JsonRedisSerializer);

        redisTemplate.afterPropertiesSet();
        return redisTemplate;
    }

}

class JodaDateTimeJsonSerializer extends JsonSerializer<DateTime> {
    @Override
    public void serialize(DateTime dateTime, JsonGenerator jsonGenerator, SerializerProvider serializerProvider) throws IOException {
        jsonGenerator.writeString(dateTime.toString("yyyy-MM-dd HH:mm:ss"));
    }
}

class JodaDateTimeJsonDeserializer extends JsonDeserializer<DateTime> {
    @Override
    public DateTime deserialize(JsonParser jsonParser, DeserializationContext deserializationContext) throws IOException, JsonProcessingException {
        String dateString = jsonParser.readValueAs(String.class);
        DateTimeFormatter dateTimeFormatter = DateTimeFormat.forPattern("yyyy-MM-dd HH:mm:ss");
        return dateTimeFormatter.parseDateTime(dateString);
    }
}

4.5 常量管理

由于代码中会用到一些共有的常量，我们应该将其抽离出来。

public class LotteryConstants {

    /**
     * 表示正在抽奖的用户标记
     */
    public final static String DRAWING = "DRAWING";
    
    /**
     * 活动标记 LOTTERY:lotteryID
     */
    public final static String LOTTERY = "LOTTERY";
    
    /**
     * 奖品数据  LOTTERY_PRIZE:lotteryID:PrizeId
     */
    public final static String LOTTERY_PRIZE = "LOTTERY_PRIZE";
    
    /**
     * 默认奖品数据  DEFAULT_LOTTERY_PRIZE:lotteryID
     */
    public final static String DEFAULT_LOTTERY_PRIZE = "DEFAULT_LOTTERY_PRIZE";

    public enum PrizeTypeEnum {
        THANK(-1), NORMAL(1), UNIQUE(2);
        private int value;

        private PrizeTypeEnum(int value) {
            this.value = value;
        }

        public int getValue() {
            return this.value;
        }
    }

    /**
     * 奖项缓存：LOTTERY_ITEM:LOTTERY_ID
     */
    public final static String LOTTERY_ITEM = "LOTTERY_ITEM";
    
    /**
     * 默认奖项： DEFAULT_LOTTERY_ITEM:LOTTERY_ID
     */
    public final static String DEFAULT_LOTTERY_ITEM = "DEFAULT_LOTTERY_ITEM";

}

public enum ReturnCodeEnum {

    SUCCESS("0000", "成功"),

    LOTTER_NOT_EXIST("9001", "指定抽奖活动不存在"),

    LOTTER_FINISH("9002", "活动已结束"),

    LOTTER_REPO_NOT_ENOUGHT("9003", "当前奖品库存不足"),

    LOTTER_ITEM_NOT_INITIAL("9004", "奖项数据未初始化"),

    LOTTER_DRAWING("9005", "上一次抽奖还未结束"),

    REQUEST_PARAM_NOT_VALID("9998", "请求参数不正确"),

    SYSTEM_ERROR("9999", "系统繁忙,请稍后重试");

    private String code;

    private String msg;

    private ReturnCodeEnum(String code, String msg) {
        this.code = code;
        this.msg = msg;
    }

    public String getCode() {
        return code;
    }

    public String getMsg() {
        return msg;
    }

    public String getCodeString() {
        return getCode() + "";
    }
}

对Redis中的key进行统一的管理。

public class RedisKeyManager {

    /**
     * 正在抽奖的key
     *
     * @param accountIp
     * @return
     */
    public static String getDrawingRedisKey(String accountIp) {
        return new StringBuilder(LotteryConstants.DRAWING).append(":").append(accountIp).toString();
    }

    /**
     * 获取抽奖活动的key
     *
     * @param id
     * @return
     */
    public static String getLotteryRedisKey(Integer id) {
        return new StringBuilder(LotteryConstants.LOTTERY).append(":").append(id).toString();
    }

    /**
     * 获取指定活动下的所有奖品数据
     *
     * @param lotteryId
     * @return
     */
    public static String getLotteryPrizeRedisKey(Integer lotteryId) {
        return new StringBuilder(LotteryConstants.LOTTERY_PRIZE).append(":").append(lotteryId).toString();
    }

    public static String getLotteryPrizeRedisKey(Integer lotteryId, Integer prizeId) {
        return new StringBuilder(LotteryConstants.LOTTERY_PRIZE).append(":").append(lotteryId).append(":").append(prizeId).toString();
    }

    public static String getDefaultLotteryPrizeRedisKey(Integer lotteryId) {
        return new StringBuilder(LotteryConstants.DEFAULT_LOTTERY_PRIZE).append(":").append(lotteryId).toString();
    }

    public static String getLotteryItemRedisKey(Integer lotteryId) {
        return new StringBuilder(LotteryConstants.LOTTERY_ITEM).append(":").append(lotteryId).toString();
    }

    public static String getDefaultLotteryItemRedisKey(Integer lotteryId) {
        return new StringBuilder(LotteryConstants.DEFAULT_LOTTERY_ITEM).append(":").append(lotteryId).toString();
    }
}

4.6 业务代码

4.6.1 抽奖接口

我们首先编写抽奖接口，根据前台传的参数查询到具体的活动，然后进行相应的操作。（当然，前端直接是写死的/lottery/1）

@GetMapping("/{id}")
public ResultResp<LotteryItemVo> doDraw(@PathVariable("id") Integer id, HttpServletRequest request) {
    String accountIp = CusAccessObjectUtil.getIpAddress(request);
    log.info("begin LotteryController.doDraw,access user {}, lotteryId,{}:", accountIp, id);
    ResultResp<LotteryItemVo> resultResp = new ResultResp<>();
    try {
        //判断当前用户上一次抽奖是否结束
        checkDrawParams(id, accountIp);

        //抽奖
        DoDrawDto dto = new DoDrawDto();
        dto.setAccountIp(accountIp);
        dto.setLotteryId(id);
        lotteryService.doDraw(dto);

        //返回结果设置
        resultResp.setCode(ReturnCodeEnum.SUCCESS.getCode());
        resultResp.setMsg(ReturnCodeEnum.SUCCESS.getMsg());
        //对象转换
        resultResp.setResult(lotteryConverter.dto2LotteryItemVo(dto));
    } catch (Exception e) {
        return ExceptionUtil.handlerException4biz(resultResp, e);
    } finally {
        //清除占位标记
        redisTemplate.delete(RedisKeyManager.getDrawingRedisKey(accountIp));
    }
    return resultResp;
}

private void checkDrawParams(Integer id, String accountIp) {
    if (null == id) {
        throw new RewardException(ReturnCodeEnum.REQUEST_PARAM_NOT_VALID.getCode(), ReturnCodeEnum.REQUEST_PARAM_NOT_VALID.getMsg());
    }
    //采用setNx命令，判断当前用户上一次抽奖是否结束
    Boolean result = redisTemplate.opsForValue().setIfAbsent(RedisKeyManager.getDrawingRedisKey(accountIp), "1", 60, TimeUnit.SECONDS);
    //如果为false，说明上一次抽奖还未结束
    if (!result) {
        throw new RewardException(ReturnCodeEnum.LOTTER_DRAWING.getCode(), ReturnCodeEnum.LOTTER_DRAWING.getMsg());
    }
}

为了避免用户重复点击抽奖，所以我们通过Redis来避免这种问题，用户每次抽奖的时候，通过setNx给用户排队并设置过期时间；如果用户点击多次抽奖，Redis设置值的时候发现该用户上次抽奖还未结束则抛出异常。

最后用户抽奖成功的话，记得清除该标记，从而用户能够继续抽奖。

4.6.2 初始化数据

从抽奖入口进来，校验成功以后则开始业务操作。

@Override
public void doDraw(DoDrawDto drawDto) throws Exception {
    RewardContext context = new RewardContext();
    LotteryItem lotteryItem = null;
    try {
        //JUC工具 需要等待线程结束之后才能运行
        CountDownLatch countDownLatch = new CountDownLatch(1);
        //判断活动有效性
        Lottery lottery = checkLottery(drawDto);
        //发布事件，用来加载指定活动的奖品信息
        applicationContext.publishEvent(new InitPrizeToRedisEvent(this, lottery.getId(), countDownLatch));
        //开始抽奖
        lotteryItem = doPlay(lottery);
        //记录奖品并扣减库存
        countDownLatch.await(); //等待奖品初始化完成
        String key = RedisKeyManager.getLotteryPrizeRedisKey(lottery.getId(), lotteryItem.getPrizeId());
        int prizeType = Integer.parseInt(redisTemplate.opsForHash().get(key, "prizeType").toString());
        context.setLottery(lottery);
        context.setLotteryItem(lotteryItem);
        context.setAccountIp(drawDto.getAccountIp());
        context.setKey(key);
        //调整库存及记录中奖信息
        AbstractRewardProcessor.rewardProcessorMap.get(prizeType).doReward(context);
    } catch (UnRewardException u) { //表示因为某些问题未中奖，返回一个默认奖项
        context.setKey(RedisKeyManager.getDefaultLotteryPrizeRedisKey(lotteryItem.getLotteryId()));
        lotteryItem = (LotteryItem) redisTemplate.opsForValue().get(RedisKeyManager.getDefaultLotteryItemRedisKey(lotteryItem.getLotteryId()));
        context.setLotteryItem(lotteryItem);
        AbstractRewardProcessor.rewardProcessorMap.get(LotteryConstants.PrizeTypeEnum.THANK.getValue()).doReward(context);
    }
    //拼接返回数据
    drawDto.setLevel(lotteryItem.getLevel());
    drawDto.setPrizeName(context.getPrizeName());
    drawDto.setPrizeId(context.getPrizeId());
}

首先我们通过CountDownLatch来保证商品初始化的顺序，关于CountDownLatch可以查看 JUC工具该文章。

然后我们需要检验一下活动的有效性，确保活动未结束。

检验活动通过后则通过ApplicationEvent 事件实现奖品数据的加载，将其存入Redis中。或者通过ApplicationRunner在程序启动时获取相关数据。我们这使用的是事件机制。ApplicationRunner 的相关代码在下文我也顺便贴出。

事件机制

public class InitPrizeToRedisEvent extends ApplicationEvent {

    private Integer lotteryId;

    private CountDownLatch countDownLatch;

    public InitPrizeToRedisEvent(Object source, Integer lotteryId, CountDownLatch countDownLatch) {
        super(source);
        this.lotteryId = lotteryId;
        this.countDownLatch = countDownLatch;
    }

    public Integer getLotteryId() {
        return lotteryId;
    }

    public void setLotteryId(Integer lotteryId) {
        this.lotteryId = lotteryId;
    }

    public CountDownLatch getCountDownLatch() {
        return countDownLatch;
    }

    public void setCountDownLatch(CountDownLatch countDownLatch) {
        this.countDownLatch = countDownLatch;
    }
}

有了事件机制，我们还需要一个监听事件，用来初始化相关数据信息。具体业务逻辑大家可以参考下代码，有相关的注释信息，主要就是将数据库中的数据添加进redis中，需要注意的是，我们为了保证原子性，是通过HASH来存储数据的，这样之后库存扣减的时候就可以通过opsForHash来保证其原子性。

当初始化奖品信息之后，则通过countDown()方法表名执行完成，业务代码中线程阻塞的地方可以继续执行了。

@Slf4j
@Component
public class InitPrizeToRedisListener implements ApplicationListener<InitPrizeToRedisEvent> {

    @Autowired
    RedisTemplate redisTemplate;

    @Autowired
    LotteryPrizeMapper lotteryPrizeMapper;

    @Autowired
    LotteryItemMapper lotteryItemMapper;

    @Override
    public void onApplicationEvent(InitPrizeToRedisEvent initPrizeToRedisEvent) {
        log.info("begin InitPrizeToRedisListener," + initPrizeToRedisEvent);
        Boolean result = redisTemplate.opsForValue().setIfAbsent(RedisKeyManager.getLotteryPrizeRedisKey(initPrizeToRedisEvent.getLotteryId()), "1");
        //已经初始化到缓存中了，不需要再次缓存
        if (!result) {
            log.info("already initial");
            initPrizeToRedisEvent.getCountDownLatch().countDown();
            return;
        }
        QueryWrapper<LotteryItem> lotteryItemQueryWrapper = new QueryWrapper<>();
        lotteryItemQueryWrapper.eq("lottery_id", initPrizeToRedisEvent.getLotteryId());
        List<LotteryItem> lotteryItems = lotteryItemMapper.selectList(lotteryItemQueryWrapper);

        //如果指定的奖品没有了，会生成一个默认的奖项
        LotteryItem defaultLotteryItem = lotteryItems.parallelStream().filter(o -> o.getDefaultItem().intValue() == 1).findFirst().orElse(null);

        Map<String, Object> lotteryItemMap = new HashMap<>(16);
        lotteryItemMap.put(RedisKeyManager.getLotteryItemRedisKey(initPrizeToRedisEvent.getLotteryId()), lotteryItems);
        lotteryItemMap.put(RedisKeyManager.getDefaultLotteryItemRedisKey(initPrizeToRedisEvent.getLotteryId()), defaultLotteryItem);
        redisTemplate.opsForValue().multiSet(lotteryItemMap);

        QueryWrapper queryWrapper = new QueryWrapper();
        queryWrapper.eq("lottery_id", initPrizeToRedisEvent.getLotteryId());
        List<LotteryPrize> lotteryPrizes = lotteryPrizeMapper.selectList(queryWrapper);

        //保存一个默认奖项
        AtomicReference<LotteryPrize> defaultPrize = new AtomicReference<>();
        lotteryPrizes.stream().forEach(lotteryPrize -> {
            if (lotteryPrize.getId().equals(defaultLotteryItem.getPrizeId())) {
                defaultPrize.set(lotteryPrize);
            }
            String key = RedisKeyManager.getLotteryPrizeRedisKey(initPrizeToRedisEvent.getLotteryId(), lotteryPrize.getId());
            setLotteryPrizeToRedis(key, lotteryPrize);
        });
        String key = RedisKeyManager.getDefaultLotteryPrizeRedisKey(initPrizeToRedisEvent.getLotteryId());
        setLotteryPrizeToRedis(key, defaultPrize.get());
        initPrizeToRedisEvent.getCountDownLatch().countDown(); //表示初始化完成
        log.info("finish InitPrizeToRedisListener," + initPrizeToRedisEvent);
    }

    private void setLotteryPrizeToRedis(String key, LotteryPrize lotteryPrize) {
        redisTemplate.setHashValueSerializer(new Jackson2JsonRedisSerializer<>(Object.class));
        redisTemplate.opsForHash().put(key, "id", lotteryPrize.getId());
        redisTemplate.opsForHash().put(key, "lotteryId", lotteryPrize.getLotteryId());
        redisTemplate.opsForHash().put(key, "prizeName", lotteryPrize.getPrizeName());
        redisTemplate.opsForHash().put(key, "prizeType", lotteryPrize.getPrizeType());
        redisTemplate.opsForHash().put(key, "totalStock", lotteryPrize.getTotalStock());
        redisTemplate.opsForHash().put(key, "validStock", lotteryPrize.getValidStock());
    }
}

上面部分是通过事件的方法来初始化数据，下面我们说下ApplicationRunner的方式：

这种方式很简单，在项目启动的时候将数据加载进去即可。

我们只需要实现ApplicationRunner接口即可，然后在run方法中从数据库读取数据加载到Redis中。

@Slf4j
@Component
public class LoadDataApplicationRunner implements ApplicationRunner {


    @Autowired
    RedisTemplate redisTemplate;

    @Autowired
    LotteryMapper lotteryMapper;

    @Override
    public void run(ApplicationArguments args) throws Exception {
        log.info("=========begin load lottery data to Redis===========");
        //加载当前抽奖活动信息
        Lottery lottery = lotteryMapper.selectById(1);

        log.info("=========finish load lottery data to Redis===========");
    }
}

4.6.3 抽奖

我们在使用事件进行数据初始化的时候，可以同时进行抽奖操作，但是注意的是这个时候需要使用countDownLatch.await();来阻塞当前线程，等待数据初始化完成。

在抽奖的过程中，我们首先尝试从Redis中获取相关数据，如果Redis中没有则从数据库中加载数据，如果数据库中也没查询到相关数据，则表明相关的数据没有配置完成。

获取数据之后，我们就该开始抽奖了。抽奖的核心在于随机性以及概率性，咱们总不能随便抽抽都能抽到一等奖吧？所以我们需要在表中设置每个奖项的概率性。如下所示：

在我们抽奖的时候需要根据概率划分处相关区间。我们可以通过Debug的方式来查看一下具体怎么划分的：

奖项的概率越大，区间越大；大家看到的顺序是不同的，由于我们在上面通过Collections.shuffle(lotteryItems);将集合打乱了，所以这里看到的不是顺序展示的。

在生成对应区间后，我们通过生成随机数，看随机数落在那个区间中，然后将对应的奖项返回。这就实现了我们的抽奖过程。

private LotteryItem doPlay(Lottery lottery) {
    LotteryItem lotteryItem = null;
    QueryWrapper<LotteryItem> queryWrapper = new QueryWrapper<>();
    queryWrapper.eq("lottery_id", lottery.getId());
    Object lotteryItemsObj = redisTemplate.opsForValue().get(RedisKeyManager.getLotteryItemRedisKey(lottery.getId()));
    List<LotteryItem> lotteryItems;
    //说明还未加载到缓存中，同步从数据库加载，并且异步将数据缓存
    if (lotteryItemsObj == null) {
        lotteryItems = lotteryItemMapper.selectList(queryWrapper);
    } else {
        lotteryItems = (List<LotteryItem>) lotteryItemsObj;
    }
    //奖项数据未配置
    if (lotteryItems.isEmpty()) {
        throw new BizException(ReturnCodeEnum.LOTTER_ITEM_NOT_INITIAL.getCode(), ReturnCodeEnum.LOTTER_ITEM_NOT_INITIAL.getMsg());
    }
    int lastScope = 0;
    Collections.shuffle(lotteryItems);
    Map<Integer, int[]> awardItemScope = new HashMap<>();
    //item.getPercent=0.05 = 5%
    for (LotteryItem item : lotteryItems) {
        int currentScope = lastScope + new BigDecimal(item.getPercent().floatValue()).multiply(new BigDecimal(mulriple)).intValue();
        awardItemScope.put(item.getId(), new int[]{lastScope + 1, currentScope});
        lastScope = currentScope;
    }
    int luckyNumber = new Random().nextInt(mulriple);
    int luckyPrizeId = 0;
    if (!awardItemScope.isEmpty()) {
        Set<Map.Entry<Integer, int[]>> set = awardItemScope.entrySet();
        for (Map.Entry<Integer, int[]> entry : set) {
            if (luckyNumber >= entry.getValue()[0] && luckyNumber <= entry.getValue()[1]) {
                luckyPrizeId = entry.getKey();
                break;
            }
        }
    }
    for (LotteryItem item : lotteryItems) {
        if (item.getId().intValue() == luckyPrizeId) {
            lotteryItem = item;
            break;
        }
    }
    return lotteryItem;
}

4.6.4 调整库存及记录

在调整库存的时候，我们需要考虑到每个奖品类型的不同，根据不同类型的奖品采取不同的措施。比如如果是一些价值高昂的奖品，我们需要通过分布式锁来确保安全性；或者比如有些商品我们需要发送相应的短信；所以我们需要采取一种具有扩展性的实现机制。

具体的实现机制可以看下方的类图，我首先定义一个奖品方法的接口（RewardProcessor），然后定义一个抽象类（AbstractRewardProcessor），抽象类中定义了模板方法，然后我们就可以根据不同的类型创建不同的处理器即可，这大大加强了我们的扩展性。

比如我们这边就创建了库存充足处理器及库存不足处理器。

接口：

public interface RewardProcessor<T> {

    void doReward(RewardContext context);

}

抽象类：

@Slf4j
public abstract class AbstractRewardProcessor implements RewardProcessor<RewardContext>, ApplicationContextAware {

    public static Map<Integer, RewardProcessor> rewardProcessorMap = new ConcurrentHashMap<Integer, RewardProcessor>();

    @Autowired
    protected RedisTemplate redisTemplate;

    private void beforeProcessor(RewardContext context) {
    }

    @Override
    public void doReward(RewardContext context) {
        beforeProcessor(context);
        processor(context);
        afterProcessor(context);
    }

    protected abstract void afterProcessor(RewardContext context);


    /**
     * 发放对应的奖品
     *
     * @param context
     */
    protected abstract void processor(RewardContext context);

    /**
     * 返回当前奖品类型
     *
     * @return
     */
    protected abstract int getAwardType();

    @Override
    public void setApplicationContext(ApplicationContext applicationContext) throws BeansException {
        rewardProcessorMap.put(LotteryConstants.PrizeTypeEnum.THANK.getValue(), (RewardProcessor) applicationContext.getBean(NoneStockRewardProcessor.class));
        rewardProcessorMap.put(LotteryConstants.PrizeTypeEnum.NORMAL.getValue(), (RewardProcessor) applicationContext.getBean(HasStockRewardProcessor.class));
    }
}

我们可以从抽象类中的doReward方法处开始查看，比如我们这边先查看库存充足处理器中的代码：

库存处理器执行的时候首相将Redis中对应的奖项库存减1，这时候是不需要加锁的，因为这个操作是原子性的。

当扣减后，我们根据返回的值判断商品库存是否充足，这个时候库存不足则提示未中奖或者返回一个默认商品。

最后我们还需要记得更新下数据库中的相关数据。

@Override
protected void processor(RewardContext context) {
    //扣减库存（redis的更新）
    Long result = redisTemplate.opsForHash().increment(context.getKey(), "validStock", -1);
    //当前奖品库存不足，提示未中奖，或者返回一个兜底的奖品
    if (result.intValue() < 0) {
        throw new UnRewardException(ReturnCodeEnum.LOTTER_REPO_NOT_ENOUGHT.getCode(), ReturnCodeEnum.LOTTER_REPO_NOT_ENOUGHT.getMsg());
    }
    List<Object> propertys = Arrays.asList("id", "prizeName");
    List<Object> prizes = redisTemplate.opsForHash().multiGet(context.getKey(), propertys);
    context.setPrizeId(Integer.parseInt(prizes.get(0).toString()));
    context.setPrizeName(prizes.get(1).toString());
    //更新库存（数据库的更新）
    lotteryPrizeMapper.updateValidStock(context.getPrizeId());
}

方法执行完成之后，我们需要执行afterProcessor方法：

这个地方我们是通过异步任务异步存入抽奖记录信息。

@Override
protected void afterProcessor(RewardContext context) {
    asyncLotteryRecordTask.saveLotteryRecord(context.getAccountIp(), context.getLotteryItem(), context.getPrizeName());
}

在这边我们可以发现是通过Async注解，指定一个线程池，开启一个异步执行的方法。

@Slf4j
@Component
public class AsyncLotteryRecordTask {

    @Autowired
    LotteryRecordMapper lotteryRecordMapper;

    @Async("lotteryServiceExecutor")
    public void saveLotteryRecord(String accountIp, LotteryItem lotteryItem, String prizeName) {
        log.info(Thread.currentThread().getName() + "---saveLotteryRecord");
        //存储中奖信息
        LotteryRecord record = new LotteryRecord();
        record.setAccountIp(accountIp);
        record.setItemId(lotteryItem.getId());
        record.setPrizeName(prizeName);
        record.setCreateTime(LocalDateTime.now());
        lotteryRecordMapper.insert(record);
    }
}

创建一个线程池：相关的配置信息是我们定义在YML文件中的数据。

@Configuration
@EnableAsync
@EnableConfigurationProperties(ThreadPoolExecutorProperties.class)
public class ThreadPoolExecutorConfig {

    @Bean(name = "lotteryServiceExecutor")
    public Executor lotteryServiceExecutor(ThreadPoolExecutorProperties poolExecutorProperties) {
        ThreadPoolTaskExecutor executor = new ThreadPoolTaskExecutor();
        executor.setCorePoolSize(poolExecutorProperties.getCorePoolSize());
        executor.setMaxPoolSize(poolExecutorProperties.getMaxPoolSize());
        executor.setQueueCapacity(poolExecutorProperties.getQueueCapacity());
        executor.setThreadNamePrefix(poolExecutorProperties.getNamePrefix());
        executor.setRejectedExecutionHandler(new ThreadPoolExecutor.CallerRunsPolicy());
        return executor;
    }
}

@Data
@ConfigurationProperties(prefix = "async.executor.thread")
public class ThreadPoolExecutorProperties {
    private int corePoolSize;
    private int maxPoolSize;
    private int queueCapacity;
    private String namePrefix;
}