Improve connection speed by running fewer commands

[chayim]

Today, connecting to a Redis instance, involves running multiple commands, and parsing the response repeatedly, this includes usage such as setinfo, which itself is run twice. The example below illustrates, that it's possible to pass all connection configuration information within a pipeline. In turn, the HELLO call returns the version of Redis, which could perhaps be used later for other things.

This is marked as good-first-issue not because it's necessarily easy to implement, inasmuch as it's a good entrypoint, to learning how connections to redis occur.

127.0.0.1:6379> multi
OK
127.0.0.1:6379(TX)> auth bob smurf
QUEUED
127.0.0.1:6379(TX)> select 5
QUEUED
127.0.0.1:6379[5](TX)> hello 3
QUEUED
127.0.0.1:6379[5](TX)> client setinfo lib-name predis
QUEUED
127.0.0.1:6379[5](TX)>  client setinfo lib-ver 55.5.0
QUEUED
127.0.0.1:6379[5](TX)> client tracking on optin
QUEUED
127.0.0.1:6379[5](TX)> CLIENT caching yes
QUEUED
127.0.0.1:6379[5](TX)> exec
1) OK
2) OK
3) 1# "server" => "redis"
   2# "version" => "7.2.3"
   3# "proto" => (integer) 3
   4# "id" => (integer) 26
   5# "mode" => "standalone"
   6# "role" => "master"
   7# "modules" => (empty array)
4) OK
5) OK
6) OK
7) OK
127.0.0.1:6379[5]> 

[fatelei]

assign to me

[github-actions]

This issue is marked stale. It will be closed in 30 days if it is not updated.