evoluxbr / python-redis-rate-limit Goto Github PK

The rate limiter should have the ability to increase usage in bulk. This will allow the rate limit to assign different weights to different requests which will make the rate limiter more versatile.

This feature entails adding a Lua script to use incrby instead of incr and being able to pass in an argument to the script that will be used to increase the count.

It makes sense, since the method returns True once the usage equals the maximum number of requests in period.

It's always a good idea to have a short and concise name when importing the module.

I'm suggesting rlimiter:

from rlimiter import RateLimit, TooManyRequests

redis_version = self._redis.info['redis_version'] is_supported = Version(str(redis_version)) >= Version('2.6.0')

This does not work on Amazon's ElastiCache as redis_version is a number, and Version() blows up with expected string or a byte-like object....

It looks as though this would work the same in py2 as py3, but it would be nice if the Readme and Setup indicated the explicit support.

Currently, a new rate limiter has to be created to have a separate limit per-client. Would you accept a PR to support per-client rate limits? I would add a check in the constructor here. If the max_requests is a dict, it will be treated as a mapping between client and max_requests for that client. Otherwise, it will be a global limit.

I'll try to implement a way so we can mock the redis.

This shouldn't require a live server for testing.

Nice!

Your library will be shared in our next RedisWeekly :)!

It would make easier to use this project with Flask and other web frameworks.

Greetings,

I see on PyPI that this project is MIT-licensed. It would be nice to update this repository to reflect the same, if it truly is licensed under the MIT license.

Thank you for a great package.