Open teburd opened 8 years ago
It might. Since this crate is just a wrapper around a C++ codebase, I'm not sure whether it would be meaningful to wrap cassandra-c++ futures in tokio futures, or not.
I would certainly use tokio if/when I work on a native rust driver.
Leaving this open for consideration, though.
It seems like it wouldn't be too much of a hassle to implement the futures-rs Future
trait for this crate's own Future
implementation -- that would at least allow for easier interop between the two systems.
@tupshin I began work on a native rust driver, but the tokio-proto crate clearly wasn't stable yet so I've put it off for now, I got the basic message encoding/decoding pieces sort of hacked together though
Tokio is coming along very quickly, and I suspect it will be ready sooner rather than later. I'll be quite interested in contributing to that effort if you decide to push it forward.
I fully intend on pushing it forward unless someone beats me to it once I see some stability with the Tokio API. It looks like that is finally happening and that the API will fit Cassandra well. Simply having the entire protocol in a manner close to what is needed by tokio is the goal I had with https://github.com/bfrog/cql-protocol while avoiding directly requiring tokio there. Then https://github.com/bfrog/tokio-cql would glue the two. I'd be more than happy to create a group to work on these projects if that sounds interesting to you. I have a strong need for that crate in the near future.
Many of the async clients/servers are moving towards the common futures/tokio stuff, might be worth integrating cassandra-rs to use the rust futures crate