Skip to main content

基础使用

官方文档

安装​

  • 同步库
# 主要是json格式的存储
pip install rejson

# 全功能的redis库
pip install redis
  • 异步库
pip install aioredis

#or 稳定版
pip install hiredis

# 2022 最新
pip install redis>=4.2.0rc1

引入​

from redis import asyncio as aioredis

基础使用​

  • 一次性链接，每次查询都会重新链接，而链接的消耗比查询的消耗大很多
import aioredis

redis = await aioredis.from_url("redis://localhost",  db=1)
redis = await aioredis.from_url("redis://localhost/1")
redis = await aioredis.from_url("redis://user:sEcRet@localhost/")
redis = await aioredis.from_url(
    url="redis://localhost", #or host="localhost"
    username="user", 
    password="sEcRet",
    encoding="utf-8",        # 指定我的输入都是 utf-8 编码
    decode_responses=True,   # 指定编码，让redis返回已经变吗的数据，默认 utf-8
    single_connection_client=True,
)

使用连接池（自动重新连接）​

# 创建连接池

使用示例

redis 同步​

# 创建连接池

aioredis 异步​

async def main():
    redis = aioredis.from_url("redis://localhost")
    await redis.set("key", "string-value")
    bin_value = await redis.get("key")
    assert bin_value == b"string-value"
    redis = aioredis.from_url("redis://localhost", decode_responses=True)
    str_value = await redis.get("key")
    assert str_value == "string-value"

    await redis.close()


if __name__ == "__main__":
    asyncio.run(main())
async def main():
    redis = aioredis.from_url("redis://localhost", decode_responses=True)

    await redis.hset("hash", mapping={"key1": "value1", "key2": "value2", "key3": 123})

    result = await redis.hgetall("hash")
    assert result == {
        "key1": "value1",
        "key2": "value2",
        "key3": "123",  # note that Redis returns int as string
    }

    await redis.close()

事务（原子性）​

async def main():
    redis = await aioredis.from_url("redis://localhost")
    async with redis.pipeline(transaction=True) as pipe:
        ok1, ok2 = await (
          pipe.set("key1", "value1").set("key2", "value2").execute()
        )
    assert ok1
    assert ok2

使用独立链接​

async def main():
    redis = aioredis.from_url(
        "redis://localhost", encoding="utf-8", decode_responses=True
    )
    async with redis.Client() as connect:
            await conn.set("my-key", "value")
        val = await conn.get('my-key')
    print(val)
    
# or
async def main_single():
    # Create a redis client with only a single connection.
    redis = aioredis.Redis(
        host="localhost",
        encoding="utf-8",
        decode_responses=True,
        single_connection_client=True,
    )
    ok = await redis.execute_command("set", "my-key", "some value")
    assert ok is True

    str_value = await redis.execute_command("get", "my-key")
    assert str_value == "some value"

    print("str value:", str_value)
    # the connection is automatically closed by GC.

使用连接池​

async def redis_pool():
    # Redis client bound to pool of connections (auto-reconnecting).
    redis = aioredis.from_url(
        "redis://localhost", encoding="utf-8", decode_responses=True
    )
    await redis.set("my-key", "value")
    val = await redis.get("my-key")
    print(val)

注意事项【坑】

链接时间慢​

发病环境

  • 系统平台：windows10
  • redis版本：6.0

临床表现

  • 每次链接时间2秒以上，使用set，get，等所以命令的首次使用的过长
  • 而且统一个redis，统一系统，采用node的redis模块进行链接则不会发生

相似病例

  • https://blog.csdn.net/qq_39091354/article/details/111934540

病因分析

  • 这个病因只会在windows平台下发生，linux下未测试

  • 如果使用 localhost 作为url给python，python主动请求连接::1这个IP地址，导致了大量的超时和连接失败，最后才会尝试使用127.0.0.1

  • 因为windows10下的localhost实际地址是 ::1，是IPv6地址

  • 因为一般redis不会在windows上使用

    Pinging DESKTOP-S9VGC7V [::1] with 32 bytes of data:
    Reply from ::1: time<1ms
    Reply from ::1: time<1ms
    Reply from ::1: time<1ms
    Reply from ::1: time<1ms

治疗方案

  • 不适用 localhost 来用作redis的链接url

扩展阅读

  • 什么是 127.0.0.1
  • 什么是 localhost
  • 什么是 ::1 - (0000:0000:0000:0000:0000:0000:0000:0001)