Github user tzulitai commented on the issue:
https://github.com/apache/flink/pull/1813
Hi,
I've tested this on a local single node installation & cluster
installation. It works and have not come across problems.
Have a few other questions on the usage of the connector (I haven't went
through the code thoroughly enough and am not an expert Redis user, so I
apologize if I may be missing things in asking them). Most of them are my own
opinion from a user's point of view:
1) as a user, I think the current way of giving secondary key for `HASH`
and `SORTED_SET` is a bit confusing. Is the current design a common usage
pattern across Redis client libraries? If not, IMO, perhaps a
`getSecondaryKeyFromData` in the `RedisMapper` interface will be more explicit.
2) I wonder what's the reason for having our own `JedisPoolConfig`, when
Jedis has its own `JedisPoolConfig`? If there's a need for it for
Flink-specific `RedisSink` settings, perhaps the name conflict is a bit
misleading?
3) Would it be better to have a `RedisDataType.PUB` instead of
`RedisDataType.PUBSUB` to better acknowledge the `publish` method on Jedis
clients? We can have `RedisDataType.SUB` when we add a Redis source to the
connector.
4) Can `RedisDataType.LIST` be changed to `RedisDataType.LPUSH` and
`RedisDataType.RPUSH`? If it isn't too hard to add, the connector would be much
more useful with this.
5) In general, I have the feeling that `RedisDataType` should be named as
`RedisCommandType`, as the `RedisMapper` is actually mapping the incoming data
to a Redis command / action.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
