Components refactor #35
Conversation
|
|
||
| [patch.crates-io] | ||
| abscissa_core = { git = "https://github.com/str4d/abscissa.git", rev = "8da8e10cad02aa4a93099cb4942b48fbdc54eb89" } | ||
| abscissa_tokio = { git = "https://github.com/str4d/abscissa.git", rev = "8da8e10cad02aa4a93099cb4942b48fbdc54eb89" } |
There was a problem hiding this comment.
This is iqlusioninc/abscissa#985.
str4d
force-pushed
the
components-refactor
branch
from
49174de to
3542cbd
Compare
|
|
||
| info!("Spawned Zallet tasks"); | ||
|
|
||
| // ongoing tasks. |
There was a problem hiding this comment.
(Not specific to this PR.)
Does pin! have a side effect causing its argument to be pinned in the current scope, or is the pin dependent on holding the returned Pin<_>? From reading the docs for std::pin, Pin<_>, and the examples for the pin! macro, I thought it was the latter.
There was a problem hiding this comment.
Oh this is tokio::pin!, not std::pin::pin!. That is sufficiently confusing (and the semantics sufficiently different) that I would prefer to see an explicit tokio::pin!(...).
|
|
||
| Ok::<_, Error>(()) | ||
| }) | ||
| .map_err(|e| FrameworkErrorKind::ComponentError.context(e))?; |
There was a problem hiding this comment.
| .map_err(|e| FrameworkErrorKind::ComponentError.context(e))?; | |
| .map_err(|e: Error| FrameworkErrorKind::ComponentError.context(e))?; |
| Ok::<(), Error>(()) | ||
| })?; | ||
|
|
||
| Ok::<_, Error>(()) |
There was a problem hiding this comment.
| Ok::<_, Error>(()) | |
| Ok(()) |
| Err(e) => Err(ErrorKind::Init.context(e)), | ||
| }?; | ||
|
|
||
| Ok::<(), Error>(()) |
There was a problem hiding this comment.
| Ok::<(), Error>(()) | |
| Ok(()) |
| info!("Creating empty database"); | ||
| } | ||
| let handle = self.handle().await?; | ||
| handle.with_mut(|mut db_data| { |
There was a problem hiding this comment.
| handle.with_mut(|mut db_data| { | |
| handle.with_mut(|mut db_data| -> Result<_, Error> { |
(Non-blocking refactoring, together with the 3 suggestions below.)
Stacked on #34.