Rob Figueiredo

If strings.Contains allocated a few bytes of memory, would everyone be happy with that? I don't think so, because it's unnecessary for the work it has to do, and it's...

I need this feature, and I'd be happy to implement it if someone could provide an outline of what to do. @achille-roussel

The use case is rebuilding some derived data store by reading the complete contents of a Topic. Presently, there is no way to know when you've reached the end of...

Yeah, I agree that now seems like the best approach. Thanks!

We have the same request. For example, we consume messages to load into ElasticSearch, which can index many more documents if you provide it a batch at a time, instead...

Looking at the current Reader implementation, it uses Batches under the covers, sending the messages in each batch one at a time over a channel. I have two alternate proposals...

@achille-roussel wondering if you have any thoughts about the API?

Tagging @alecthomas and @dominicbarnes as two recently-active maintainers to provide some feedback on the API design.

Hi @achille-roussel , The ConsumerGroup API does not really do what I wanted to. It does not expose the batch of messages, only one message at a time. You can...

I have the benchmark results, but not the code. I'm not sure I really compared a ConsumerGroup vs a single-partition Reader, though. Anyway, I found that using Conn.ReadBatch, I could...