StackExchange.Redis 模糊查询和删除[转]
作者:互联网
初始化连接对象
_connectionString = ConfigurationManager.ConnectionStrings["RedisConnectionString"].ConnectionString; _connMultiplexer = ConnectionMultiplexer.Connect(_connectionString); _db = _connMultiplexer.GetDatabase(db);
通过 keys或scan查找出匹配的key,StackExchange.Redis 中调用 keys/scan/size/flush/save/集群 等命令须使用 IServer 对象
//默认一个服务器 _server = _connMultiplexer.GetServer(_connMultiplexer.GetEndPoints()[0]); var pattern = "BUIK_201710*"; //StackExchange.Redis 会根据redis版本决定用keys还是scan(>2.8) var keys = _server.Keys(database: _db.Database, pattern: pattern); //删除一组key return _db.KeyDelete(keys.ToArray());
如果数据量很大(比如几十万个key),以上操作会很慢,甚至超时,原因未知,因为通过命令执行keys是很快的
为了提高效率,可以通过Lua脚本进行模糊查询的批量操作
var pattern = "BUIK_201710*"; var redisResult = await _db.ScriptEvaluateAsync(LuaScript.Prepare( //Redis的keys模糊查询: " local res = redis.call('KEYS', @keypattern) " + " return res "), new { @keypattern = pattern }); if (!redisResult.IsNull) { _db.KeyDelete((string[])redisResult); //删除一组key }
当然也可以通过脚本执行删除参考:
StackExchange.Redis加载Lua脚本进行模糊查询的批量删除和修改
前言
使用StackExchange.Redis没有直接相关的方法进行模糊查询的批量删除和修改操作,虽然可以通过Scan相关的方法进行模糊查询,例如:HashScan("hashkey", "*key*"),然后再使用相关的方法进行相关的批量操作,但是如果缓存数据量比较大,效率低下,那么可以使用Lua脚本进行模糊查询的批量操作:ScriptEvaluate(LuaScript.Prepare(...))。
通过keys进行模糊查询后的批量操作
批量删除
1 2 3 4 5 6 7 8 9 10 |
var redis = ConnectionMultiplexer.Connect( "127.0.0.1:6379,allowAdmin = true" );
redis.GetDatabase().ScriptEvaluate(LuaScript.Prepare(
//Redis的keys模糊查询:
" local ks = redis.call('KEYS', @keypattern) " + //local ks为定义一个局部变量,其中用于存储获取到的keys
" for i=1,#ks,5000 do " + //#ks为ks集合的个数, 语句的意思: for(int i = 1; i <= ks.Count; i+=5000)
" redis.call('del', unpack(ks, i, math.min(i+4999, #ks))) " + //Lua集合索引值从1为起始,unpack为解包,获取ks集合中的数据,每次5000,然后执行删除
" end " +
" return true "
),
new { keypattern = "mykey*" });
|
批量修改
1 2 3 4 5 6 7 |
redis.GetDatabase().ScriptEvaluate(LuaScript.Prepare(
" local ks = redis.call('KEYS', @keypattern) " +
" for i=1,#ks do " +
" redis.call('set', ks[i], @value) " +
" end " +
" return true " ),
new { keypattern = "mykey*" , value = "setval" });
|
对Hash集合下的key进行模糊查询后的批量操作
批量删除
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
redis.GetDatabase().ScriptEvaluate(LuaScript.Prepare(
" local ks = redis.call('hkeys', @hashid) " +
" local fkeys = {} " +
" for i=1,#ks do " +
//使用string.find进行匹配操作
" if string.find(ks[i], @keypattern) then " +
" fkeys[#fkeys + 1] = ks[i] " +
" end " +
" end " +
" for i=1,#fkeys,5000 do " +
" redis.call('hdel', @hashid, unpack(fkeys, i, math.min(i+4999, #fkeys))) " +
" end " +
" return true "
),
new { hashid = "hkey" , keypattern = "^mykey" }); //keypattern为可使用正则表达式
|
批量修改
1 2 3 4 5 6 7 8 9 10 11 12 13 14 |
redis.GetDatabase().ScriptEvaluate(LuaScript.Prepare(
" local ks = redis.call('hkeys', @hashid) " +
" local fkeys = {} " +
" for i=1,#ks do " +
" if string.find(ks[i], @keypattern) then " +
" fkeys[#fkeys + 1] = ks[i] " +
" end " +
" end " +
" for i=1,#fkeys do " +
" redis.call('hset', @hashid, fkeys[i], @value) " +
" end " +
" return true "
),
new { hashid = "hkey" , keypattern = "^key" , value = "hashValue" }); //keypattern为可使用正则表达式
|
对Set集合下的值进行模糊查询后的批量操作
批量删除
1 2 3 4 5 6 7 8 9 10 11 12 13 14 |
redis.GetDatabase().ScriptEvaluate(LuaScript.Prepare(
" local ks = redis.call('smembers', @keyid) " +
" local fkeys = {} " +
" for i=1,#ks do " +
" if string.find(ks[i], @keypattern) then " +
" fkeys[#fkeys + 1] = ks[i] " +
" end " +
" end " +
" for i=1,#fkeys,5000 do " +
" redis.call('srem', @keyid, unpack(fkeys, i, math.min(i+4999, #fkeys))) " +
" end " +
" return true "
),
new { keyid = "setkey" , keypattern = "^myval" }); //keypattern为可使用正则表达式
|
注意
从 Redis 2.6.0 版本开始,才可通过内置的 Lua 解释器,使用 EVAL 命令对 Lua 脚本进行求值。
转自:https://lesg.cn/Article-105088.html
标签:end,Redis,redis,fkeys,ks,call,StackExchange,查询,keypattern 来源: https://www.cnblogs.com/aaronwus/p/14288322.html