houqp commented on a change in pull request #811:
URL: https://github.com/apache/arrow-datafusion/pull/811#discussion_r693222812
##########
File path: datafusion/Cargo.toml
##########
@@ -56,9 +56,9 @@ paste = "^1.0"
num_cpus = "1.13.0"
chrono = "0.4"
async-trait = "0.1.41"
-futures = "0.3"
+futures = { version = "0.3", features = ["executor"] }
Review comment:
> The alternate compromise, which you have partly implemented in this
PR, is to implement both async and non async versions. This is similar to the
approach in the C/C++ filesystem api (props to @nealrichardson for the
pointer), which has both having synchronous and asynchronous APIs.
How about this alternative to reduce the scope of this PR? i.e. implement
both sync and async, but only use sync API to migrate existing code to the new
IO abstraction, then work on async propagation as a fast follow up.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
