基于Stream的消息队列
stream是一种数据类型,可以实现一个功能非常完善的消息队列
key:队列名称
nomkstream:如果队列不存在是否自动创建,默认创建
maxlen/minid:设置消息队列的最大消息数量
*|ID 唯一id:时间戳-递增数字
field value:消息体:键值对
XREAD 命令特点:消息可回溯,一个消息可以被多个消费者读取,可以阻塞读取,有消息漏读风险
消费者组
特点:小费可回溯,消费者争抢消息,加快消费速度,可以阻塞读取,没有漏读风险,消息确认机制。
实现:
创建一个消费者组
XGROUP CREATE stream.orders g1 0 MKSTREAM //创建队列和消费者组
最终版秒杀代码
@Slf4j
@Service
public class VoucherOrderServiceImpl extends ServiceImpl<VoucherOrderMapper, VoucherOrder> implements IVoucherOrderService {@Resourceprivate ISeckillVoucherService seckillVoucherService;@Resourceprivate RedisIdWorker redisIdWorker;@Resourceprivate RedissonClient redissonClient;@Resourceprivate StringRedisTemplate stringRedisTemplate;private static final DefaultRedisScript<Long> SECKILL_SCRIPT;static {SECKILL_SCRIPT = new DefaultRedisScript<>();SECKILL_SCRIPT.setLocation(new ClassPathResource("seckill.lua"));SECKILL_SCRIPT.setResultType(Long.class);}private static final ExecutorService SECKILL_ORDER_EXECUTOR = Executors.newSingleThreadExecutor();@PostConstructprivate void init() {SECKILL_ORDER_EXECUTOR.submit(new VoucherOrderHandler());}private class VoucherOrderHandler implements Runnable {@Overridepublic void run() {while (true) {try {// 1.获取消息队列中的订单信息 XREADGROUP GROUP g1 c1 COUNT 1 BLOCK 2000 STREAMS s1 >List<MapRecord<String, Object, Object>> list = stringRedisTemplate.opsForStream().read(Consumer.from("g1", "c1"), // 组名和消费者名StreamReadOptions.empty().count(1).block(Duration.ofSeconds(2)), //创建空的,读一个,阻塞两秒钟StreamOffset.create("stream.orders", ReadOffset.lastConsumed()) // 消息队列的名字和读取标识);// 2.判断订单信息是否为空if (list == null || list.isEmpty()) {// 如果为null,说明没有消息,继续下一次循环continue;}// 解析数据MapRecord<String, Object, Object> record = list.get(0);Map<Object, Object> value = record.getValue();VoucherOrder voucherOrder = BeanUtil.fillBeanWithMap(value, new VoucherOrder(), true);// 3.创建订单createVoucherOrder(voucherOrder);// 4.确认消息 XACKstringRedisTemplate.opsForStream().acknowledge("s1", "g1", record.getId());} catch (Exception e) {log.error("处理订单异常", e);handlePendingList();}}}private void handlePendingList() {while (true) {try {// 1.获取pending-list中的订单信息 XREADGROUP GROUP g1 c1 COUNT 1 BLOCK 2000 STREAMS s1 0List<MapRecord<String, Object, Object>> list = stringRedisTemplate.opsForStream().read(Consumer.from("g1", "c1"),StreamReadOptions.empty().count(1),StreamOffset.create("stream.orders", ReadOffset.from("0")));// 2.判断订单信息是否为空if (list == null || list.isEmpty()) {// 如果为null,说明没有异常消息,结束循环break;}// 解析数据MapRecord<String, Object, Object> record = list.get(0);Map<Object, Object> value = record.getValue();VoucherOrder voucherOrder = BeanUtil.fillBeanWithMap(value, new VoucherOrder(), true);// 3.创建订单createVoucherOrder(voucherOrder);// 4.确认消息 XACKstringRedisTemplate.opsForStream().acknowledge("s1", "g1", record.getId());} catch (Exception e) {log.error("处理订单异常", e);}}}}private void createVoucherOrder(VoucherOrder voucherOrder) {Long userId = voucherOrder.getUserId();Long voucherId = voucherOrder.getVoucherId();// 创建锁对象RLock redisLock = redissonClient.getLock("lock:order:" + userId);// 尝试获取锁boolean isLock = redisLock.tryLock();// 判断if (!isLock) {// 获取锁失败,直接返回失败或者重试log.error("不允许重复下单!");return;}try {// 5.1.查询订单int count = query().eq("user_id", userId).eq("voucher_id", voucherId).count();// 5.2.判断是否存在if (count > 0) {// 用户已经购买过了log.error("不允许重复下单!");return;}// 6.扣减库存boolean success = seckillVoucherService.update().setSql("stock = stock - 1") // set stock = stock - 1.eq("voucher_id", voucherId).gt("stock", 0) // where id = ? and stock > 0.update();if (!success) {// 扣减失败log.error("库存不足!");return;}// 7.创建订单save(voucherOrder);} finally {// 释放锁redisLock.unlock();}}@Overridepublic Result seckillVoucher(Long voucherId) {// 获取用户Long userId = UserHolder.getUser().getId();// 获取订单idlong orderId = redisIdWorker.nextId("order");// 1.执行lua脚本Long result = stringRedisTemplate.execute(SECKILL_SCRIPT,Collections.emptyList(),voucherId.toString(), userId.toString(), String.valueOf(orderId));int r = result.intValue();// 2.判断结果是否为0if (r != 0) {// 2.1.不为0 ,代表没有购买资格return Result.fail(r == 1 ? "库存不足" : "不能重复下单");}// 3.返回订单idreturn Result.ok(orderId);}
}
-- 1.参数列表
-- 1.1.优惠券id
local voucherId = ARGV[1]
-- 1.2.用户id
local userId = ARGV[2]
-- 1.3.订单id
local orderId = ARGV[3]-- 2.数据key
-- 2.1.库存key
local stockKey = 'seckill:stock:' .. voucherId
-- 2.2.订单key
local orderKey = 'seckill:order:' .. voucherId-- 3.脚本业务
-- 3.1.判断库存是否充足 get stockKey
if(tonumber(redis.call('get', stockKey)) <= 0) then-- 3.2.库存不足,返回1return 1
end
-- 3.2.判断用户是否下单 SISMEMBER orderKey userId
if(redis.call('sismember', orderKey, userId) == 1) then-- 3.3.存在,说明是重复下单,返回2return 2
end
-- 3.4.扣库存 incrby stockKey -1
redis.call('incrby', stockKey, -1)
-- 3.5.下单(保存用户)sadd orderKey userId
redis.call('sadd', orderKey, userId)
-- 3.6.发送消息到队列中, XADD stream.orders * k1 v1 k2 v2 ...
redis.call('xadd', 'stream.orders', '*', 'userId', userId, 'voucherId', voucherId, 'id', orderId)
return 0
探店达人
完善点赞功能:同一个用户只能点赞一次再点就取消,如果点赞会高亮
@Service
public class BlogServiceImpl extends ServiceImpl<BlogMapper, Blog> implements IBlogService {@Resourceprivate IUserService userService;@Resourceprivate StringRedisTemplate stringRedisTemplate;@Resourceprivate IFollowService followService;@Overridepublic Result queryHotBlog(Integer current) {// 根据用户查询Page<Blog> page = query().orderByDesc("liked").page(new Page<>(current, SystemConstants.MAX_PAGE_SIZE));// 获取当前页数据List<Blog> records = page.getRecords();// 查询用户records.forEach(blog -> {this.queryBlogUser(blog);this.isBlogLiked(blog);});return Result.ok(records);}@Overridepublic Result queryBlogById(Long id) {// 1.查询blogBlog blog = getById(id);if (blog == null) {return Result.fail("笔记不存在!");}// 2.查询blog有关的用户queryBlogUser(blog);// 3.查询blog是否被点赞isBlogLiked(blog);return Result.ok(blog);}private void isBlogLiked(Blog blog) {// 1.获取登录用户UserDTO user = UserHolder.getUser();if (user == null) {// 用户未登录,无需查询是否点赞return;}Long userId = user.getId();// 2.判断当前登录用户是否已经点赞String key = "blog:liked:" + blog.getId();Double score = stringRedisTemplate.opsForZSet().score(key, userId.toString());blog.setIsLike(score != null);}@Overridepublic Result likeBlog(Long id) {// 1.获取登录用户Long userId = UserHolder.getUser().getId();// 2.判断当前登录用户是否已经点赞String key = BLOG_LIKED_KEY + id;Double score = stringRedisTemplate.opsForZSet().score(key, userId.toString());if (score == null) {// 3.如果未点赞,可以点赞// 3.1.数据库点赞数 + 1boolean isSuccess = update().setSql("liked = liked + 1").eq("id", id).update();// 3.2.保存用户到Redis的set集合 zadd key value scoreif (isSuccess) {stringRedisTemplate.opsForZSet().add(key, userId.toString(), System.currentTimeMillis());}} else {// 4.如果已点赞,取消点赞// 4.1.数据库点赞数 -1boolean isSuccess = update().setSql("liked = liked - 1").eq("id", id).update();// 4.2.把用户从Redis的set集合移除if (isSuccess) {stringRedisTemplate.opsForZSet().remove(key, userId.toString());}}return Result.ok();}@Overridepublic Result queryBlogLikes(Long id) {String key = BLOG_LIKED_KEY + id;// 1.查询top5的点赞用户 zrange key 0 4Set<String> top5 = stringRedisTemplate.opsForZSet().range(key, 0, 4);if (top5 == null || top5.isEmpty()) {return Result.ok(Collections.emptyList());}// 2.解析出其中的用户idList<Long> ids = top5.stream().map(Long::valueOf).collect(Collectors.toList());String idStr = StrUtil.join(",", ids);// 3.根据用户id查询用户 WHERE id IN ( 5 , 1 ) ORDER BY FIELD(id, 5, 1)// mybatis plus提供自定义查询List<UserDTO> userDTOS = userService.query().in("id", ids).last("ORDER BY FIELD(id," + idStr + ")").list().stream().map(user -> BeanUtil.copyProperties(user, UserDTO.class)).collect(Collectors.toList());// 4.返回return Result.ok(userDTOS);}@Overridepublic Result saveBlog(Blog blog) {// 1.获取登录用户UserDTO user = UserHolder.getUser();blog.setUserId(user.getId());// 2.保存探店笔记boolean isSuccess = save(blog);if(!isSuccess){return Result.fail("新增笔记失败!");}// 3.查询笔记作者的所有粉丝 select * from tb_follow where follow_user_id = ?List<Follow> follows = followService.query().eq("follow_user_id", user.getId()).list();// 4.推送笔记id给所有粉丝for (Follow follow : follows) {// 4.1.获取粉丝idLong userId = follow.getUserId();// 4.2.推送String key = FEED_KEY + userId;stringRedisTemplate.opsForZSet().add(key, blog.getId().toString(), System.currentTimeMillis());}// 5.返回idreturn Result.ok(blog.getId());}@Overridepublic Result queryBlogOfFollow(Long max, Integer offset) {// 1.获取当前用户Long userId = UserHolder.getUser().getId();// 2.查询收件箱 ZREVRANGEBYSCORE key Max Min LIMIT offset countString key = FEED_KEY + userId;Set<ZSetOperations.TypedTuple<String>> typedTuples = stringRedisTemplate.opsForZSet().reverseRangeByScoreWithScores(key, 0, max, offset, 2);// 3.非空判断if (typedTuples == null || typedTuples.isEmpty()) {return Result.ok();}// 4.解析数据:blogId、minTime(时间戳)、offsetList<Long> ids = new ArrayList<>(typedTuples.size());long minTime = 0; // 2int os = 1; // 2for (ZSetOperations.TypedTuple<String> tuple : typedTuples) { // 5 4 4 2 2// 4.1.获取idids.add(Long.valueOf(tuple.getValue()));// 4.2.获取分数(时间戳)long time = tuple.getScore().longValue();if(time == minTime){os++;}else{minTime = time;os = 1;}}// 5.根据id查询blogString idStr = StrUtil.join(",", ids);List<Blog> blogs = query().in("id", ids).last("ORDER BY FIELD(id," + idStr + ")").list();for (Blog blog : blogs) {// 5.1.查询blog有关的用户queryBlogUser(blog);// 5.2.查询blog是否被点赞isBlogLiked(blog);}// 6.封装并返回ScrollResult r = new ScrollResult();r.setList(blogs);r.setOffset(os);r.setMinTime(minTime);return Result.ok(r);}private void queryBlogUser(Blog blog) {Long userId = blog.getUserId();User user = userService.getById(userId);blog.setName(user.getNickName());blog.setIcon(user.getIcon());}
}
好友关注
共同关注:查交集(Set)
@Service
public class FollowServiceImpl extends ServiceImpl<FollowMapper, Follow> implements IFollowService {@Resourceprivate StringRedisTemplate stringRedisTemplate;@Resourceprivate IUserService userService;@Overridepublic Result follow(Long followUserId, Boolean isFollow) {// 1.获取登录用户Long userId = UserHolder.getUser().getId();String key = "follows:" + userId;// 1.判断到底是关注还是取关if (isFollow) {// 2.关注,新增数据Follow follow = new Follow();follow.setUserId(userId);follow.setFollowUserId(followUserId);boolean isSuccess = save(follow);if (isSuccess) {// 把关注用户的id,放入redis的set集合 sadd userId followerUserIdstringRedisTemplate.opsForSet().add(key, followUserId.toString());}} else {// 3.取关,删除 delete from tb_follow where user_id = ? and follow_user_id = ?boolean isSuccess = remove(new QueryWrapper<Follow>().eq("user_id", userId).eq("follow_user_id", followUserId));if (isSuccess) {// 把关注用户的id从Redis集合中移除stringRedisTemplate.opsForSet().remove(key, followUserId.toString());}}return Result.ok();}@Overridepublic Result isFollow(Long followUserId) {// 1.获取登录用户Long userId = UserHolder.getUser().getId();// 2.查询是否关注 select count(*) from tb_follow where user_id = ? and follow_user_id = ?Integer count = query().eq("user_id", userId).eq("follow_user_id", followUserId).count();// 3.判断return Result.ok(count > 0);}@Overridepublic Result followCommons(Long id) {// 1.获取当前用户Long userId = UserHolder.getUser().getId();String key = "follows:" + userId;// 2.求交集String key2 = "follows:" + id;Set<String> intersect = stringRedisTemplate.opsForSet().intersect(key, key2);if (intersect == null || intersect.isEmpty()) {// 无交集return Result.ok(Collections.emptyList());}// 3.解析id集合List<Long> ids = intersect.stream().map(Long::valueOf).collect(Collectors.toList());// 4.查询用户List<UserDTO> users = userService.listByIds(ids).stream().map(user -> BeanUtil.copyProperties(user, UserDTO.class)).collect(Collectors.toList());return Result.ok(users);}
}
关注推送
拉模式:读扩散:每次都要获取消息然后根据时间戳进行排序,延迟
推模式:写扩散:要把消息写给所有人,内存占用高
推拉结合模式:读写混合:
普通人发消息,由于粉丝比较少,所以直接推模式塞进去
大V发消息,首先要区分普通粉丝和活跃粉丝,对于普通粉丝采用拉模式,活跃粉丝推模式
本项目采用推模式!
List里面只能角标查询,而Sorted支持按照score值范围查询,支持滚动分页
推送到粉丝收件箱
@Overridepublic Result saveBlog(Blog blog) {// 1.获取登录用户UserDTO user = UserHolder.getUser();blog.setUserId(user.getId());// 2.保存探店笔记boolean isSuccess = save(blog);if(!isSuccess){return Result.fail("新增笔记失败!");}// 3.查询笔记作者的所有粉丝 select * from tb_follow where follow_user_id = ?List<Follow> follows = followService.query().eq("follow_user_id", user.getId()).list();// 4.推送笔记id给所有粉丝for (Follow follow : follows) {// 4.1.获取粉丝idLong userId = follow.getUserId();// 4.2.推送String key = FEED_KEY + userId;stringRedisTemplate.opsForZSet().add(key, blog.getId().toString(), System.currentTimeMillis());}// 5.返回idreturn Result.ok(blog.getId());}
滚动分页
滚动分页查询参数:
max:第一次查询时为当前时间戳 | 上一次查询的最小时间戳
min:0
offset:第一次直接0 | 在上一次的结果中,与最小值一样的元素的个数
count:3(查询条数)
@Data
public class ScrollResult {private List<?> list;private Long minTime;private Integer offset;
}
@Overridepublic Result queryBlogOfFollow(Long max, Integer offset) {// 1.获取当前用户Long userId = UserHolder.getUser().getId();// 2.查询收件箱 ZREVRANGEBYSCORE key Max Min LIMIT offset countString key = FEED_KEY + userId;// 滚动分页查询Set<ZSetOperations.TypedTuple<String>> typedTuples = stringRedisTemplate.opsForZSet().reverseRangeByScoreWithScores(key, 0, max, offset, 2);// 3.非空判断if (typedTuples == null || typedTuples.isEmpty()) {return Result.ok();}// 4.解析数据:blogId、minTime(时间戳)、offsetList<Long> ids = new ArrayList<>(typedTuples.size());long minTime = 0; // 2// 判断重复int os = 1; // 2for (ZSetOperations.TypedTuple<String> tuple : typedTuples) { // 5 4 4 2 2// 4.1.获取idids.add(Long.valueOf(tuple.getValue()));// 4.2.获取分数(时间戳)long time = tuple.getScore().longValue();if(time == minTime){os++;}else{minTime = time;os = 1;}}// 5.根据id查询blog 不能用listByids,要用orderbyString idStr = StrUtil.join(",", ids);List<Blog> blogs = query().in("id", ids).last("ORDER BY FIELD(id," + idStr + ")").list();for (Blog blog : blogs) {// 5.1.查询blog有关的用户queryBlogUser(blog);// 5.2.查询blog是否被点赞isBlogLiked(blog);}// 6.封装并返回ScrollResult r = new ScrollResult();r.setList(blogs);r.setOffset(os);r.setMinTime(minTime);return Result.ok(r);}
附近商户
导入店铺数据
@Testvoid loadShopData() {// 1.查询店铺信息List<Shop> list = shopService.list();// 2.把店铺分组,按照typeId分组,typeId一致的放到一个集合Map<Long, List<Shop>> map = list.stream().collect(Collectors.groupingBy(Shop::getTypeId));// 3.分批完成写入Redisfor (Map.Entry<Long, List<Shop>> entry : map.entrySet()) {// 3.1.获取类型idLong typeId = entry.getKey();String key = SHOP_GEO_KEY + typeId;// 3.2.获取同类型的店铺的集合List<Shop> value = entry.getValue();List<RedisGeoCommands.GeoLocation<String>> locations = new ArrayList<>(value.size());// 3.3.写入redis GEOADD key 经度 纬度 memberfor (Shop shop : value) {// stringRedisTemplate.opsForGeo().add(key, new Point(shop.getX(), shop.getY()), shop.getId().toString());locations.add(new RedisGeoCommands.GeoLocation<>(shop.getId().toString(),new Point(shop.getX(), shop.getY())));}stringRedisTemplate.opsForGeo().add(key, locations);}}
public Result queryShopByType(Integer typeId, Integer current, Double x, Double y) {// 1.判断是否需要根据坐标查询if (x == null || y == null) {// 不需要坐标查询,按数据库查询Page<Shop> page = query().eq("type_id", typeId).page(new Page<>(current, SystemConstants.DEFAULT_PAGE_SIZE));// 返回数据return Result.ok(page.getRecords());}// 2.计算分页参数int from = (current - 1) * SystemConstants.DEFAULT_PAGE_SIZE;int end = current * SystemConstants.DEFAULT_PAGE_SIZE;// 3.查询redis、按照距离排序、分页。结果:shopId、distanceString key = SHOP_GEO_KEY + typeId;GeoResults<RedisGeoCommands.GeoLocation<String>> results = stringRedisTemplate.opsForGeo() // GEOSEARCH key BYLONLAT x y BYRADIUS 10 WITHDISTANCE.search(key,GeoReference.fromCoordinate(x, y),new Distance(5000),RedisGeoCommands.GeoSearchCommandArgs.newGeoSearchArgs().includeDistance().limit(end));// 4.解析出idif (results == null) {return Result.ok(Collections.emptyList());}List<GeoResult<RedisGeoCommands.GeoLocation<String>>> list = results.getContent();if (list.size() <= from) {// 没有下一页了,结束return Result.ok(Collections.emptyList());}// 4.1.截取 from ~ end的部分List<Long> ids = new ArrayList<>(list.size());Map<String, Distance> distanceMap = new HashMap<>(list.size());list.stream().skip(from).forEach(result -> {// 4.2.获取店铺idString shopIdStr = result.getContent().getName();ids.add(Long.valueOf(shopIdStr));// 4.3.获取距离Distance distance = result.getDistance();distanceMap.put(shopIdStr, distance);});// 5.根据id查询ShopString idStr = StrUtil.join(",", ids);List<Shop> shops = query().in("id", ids).last("ORDER BY FIELD(id," + idStr + ")").list();for (Shop shop : shops) {shop.setDistance(distanceMap.get(shop.getId().toString()).getValue());}// 6.返回return Result.ok(shops);}
}
用户签到
位图BitMap:
@Overridepublic Result sign() {// 1.获取当前登录用户Long userId = UserHolder.getUser().getId();// 2.获取日期LocalDateTime now = LocalDateTime.now();// 3.拼接keyString keySuffix = now.format(DateTimeFormatter.ofPattern(":yyyyMM"));String key = USER_SIGN_KEY + userId + keySuffix;// 4.获取今天是本月的第几天int dayOfMonth = now.getDayOfMonth();// 5.写入Redis SETBIT key offset 1stringRedisTemplate.opsForValue().setBit(key, dayOfMonth - 1, true);return Result.ok();}
签到统计
@Overridepublic Result signCount() {// 1.获取当前登录用户Long userId = UserHolder.getUser().getId();// 2.获取日期LocalDateTime now = LocalDateTime.now();// 3.拼接keyString keySuffix = now.format(DateTimeFormatter.ofPattern(":yyyyMM"));String key = USER_SIGN_KEY + userId + keySuffix;// 4.获取今天是本月的第几天int dayOfMonth = now.getDayOfMonth();// 5.获取本月截止今天为止的所有的签到记录,返回的是一个十进制的数字 BITFIELD sign:5:202203 GET u14 0List<Long> result = stringRedisTemplate.opsForValue().bitField(key,BitFieldSubCommands.create().get(BitFieldSubCommands.BitFieldType.unsigned(dayOfMonth)).valueAt(0));if (result == null || result.isEmpty()) {// 没有任何签到结果return Result.ok(0);}Long num = result.get(0);if (num == null || num == 0) {return Result.ok(0);}// 6.循环遍历int count = 0;while (true) {// 6.1.让这个数字与1做与运算,得到数字的最后一个bit位 // 判断这个bit位是否为0if ((num & 1) == 0) {// 如果为0,说明未签到,结束break;}else {// 如果不为0,说明已签到,计数器+1count++;}// 把数字右移一位,抛弃最后一个bit位,继续下一个bit位num >>>= 1;}return Result.ok(count);}
VU统计
@Testvoid testHyperLogLog() {String[] values = new String[1000];int j = 0;for (int i = 0; i < 1000000; i++) {j = i % 1000;values[j] = "user_" + i;if(j == 999){// 发送到RedisstringRedisTemplate.opsForHyperLogLog().add("hl2", values);}}// 统计数量Long count = stringRedisTemplate.opsForHyperLogLog().size("hl2");System.out.println("count = " + count);}