Tokio 1.0 #301
Tokio 1.0 #301
Conversation
|
Weird, it looks like |
|
Yeah, it looks like it can receive one message, and then poll_read_ready never wakes it up again. |
coolreader18
force-pushed
the
tokio1.0
branch
from
1b6853b
to
7e4bb17
Compare
|
Figured it out, I ran into the same issue as tokio-rs/tokio#3208 (comment) |
|
Hi and thanks for working on this! |
|
Oh, yeah, this is pretty much ready to merge if it passes review. |
There was a problem hiding this comment.
Ok! So I admit not knowing the inner details of Tokio and trusting that you do the right thing (or that other people will report bugs if you don't...).
So I think this is okay for merging given the two comments above are fixed/commented + that you squash all patches into a single commit.
Good work!
dbus-tokio/Cargo.toml
Outdated
| libc = "0.2.69" | ||
| tokio = {version = "0.2.4", features=["time", "macros", "io-driver", "rt-core", "rt-util", "rt-threaded", "stream"]} | ||
| tokio = {version = "1.0", features=["time", "net"]} | ||
| parking_lot = "0.11" |
There was a problem hiding this comment.
Unless there is a strong reason not to use std's Mutex, I would prefer to avoid additional dependencies.
dbus-tokio/src/connection.rs
Outdated
| inner: ResourceErrorKind | ||
| } | ||
| #[derive(Debug)] | ||
| enum ResourceErrorKind { |
There was a problem hiding this comment.
How about we make the enum public and skip "IOResourceError"? Potentially with an added non_exhaustive attribute.
There was a problem hiding this comment.
Alright, wasn't sure whether I should add more public API surface area, since Box<dyn Error> is pretty anonymous, but I guess it makes sense that a dbus IO reactor would fail with either an IO or a dbus error :)
There was a problem hiding this comment.
Oh, neither of the inner error types implement Clone.
|
And it's in! Thanks! |
Fixes #289
Currently it's compiling and I think it's roughly correct, but it's timing out while waiting for a response, even though the last call to poll_internal comes because read is ready.