redis-objects
redis-objects copied to clipboard
nateware
Map Redis types directly to Ruby objects
Based on https://github.com/nateware/redis-objects/pull/213 and the comments on https://github.com/nateware/redis-objects/commit/4854553f451c1b983ce620e205c45acea1fefc5b, the current README incorrectly describes an issue with the `expiration` setting when it is actually meant to be describing an issue with...
I encountered the same issue as https://github.com/nateware/redis-objects/pull/157 solved, but is for `lock` method. I am not sure why https://github.com/nateware/redis-objects/pull/157 didn't let `lock` to use `redis_options`, but I think it should...
According to [the documentation](https://redis.io/commands/setnx), you shouldn't use SETNX for locking and instead use the [Redlock algorithm as described here](https://redis.io/topics/distlock). Since there's already a [redlock-rb](https://github.com/antirez/redlock-rb) gem, integrating it shouldn't be too...
Imagine a base class called service ``` class Service < ActiveRecord::Base include Redis::Objects lock :mylock, :expiration => 15.minutes, :timeout => 10.seconds end ``` And two subclasses Service 1 ``` class...
Up until redis-objects 0.8.0, it was possible to do something like the following so one could atomically find and remove elements from a SortedSet with a rangebyscore. ``` require 'redis'...
In certain situation, I need to use multi to ensure atomic of a block of command. Current implementation of Counter#incrbyfloat include a **too smart** .to_f which make this impracticle.
I am aware of the change made in this PR: https://github.com/nateware/redis-objects/pull/68, although dont really understand why to keep `0` as the return value of calling `sorted_set[non_existent_member]`. It is a bit...
It would be great to have the possibility to use :aggregate and :weight options with interstore and unionstore commands.
