The complexity of OpsQueue in terms of the work being done is not particularly high, OpsQueue does not have many database tables, it performs some queries on these tables and provides a HTTP API and Rust/Python FFI layers.
However the complexity of some of the code that implements this work is quite high, the issue being that this raises the barrier to new contributors and increases time needed to implement new features or maintain existing ones. For example when reading the following code a new (or existing) contributor may begin to scratch their head 😃
async fn transaction<O, E, F>(&mut self, f: F) -> Result<O, E>
where
for<'t> F: FnOnce(Conn<Self::Writable, Tx<'t, '_>>) -> BoxFuture<'t, Result<O, E>>
+ Send
+ Sync
+ 't,
O: Send,
E: From<sqlx::Error> + Send,
The purpose of this issue is as a place to discuss whether it would be possible to simplify/replace any aspects of the code in relation to the async usage.
The complexity of OpsQueue in terms of the work being done is not particularly high, OpsQueue does not have many database tables, it performs some queries on these tables and provides a HTTP API and Rust/Python FFI layers.
However the complexity of some of the code that implements this work is quite high, the issue being that this raises the barrier to new contributors and increases time needed to implement new features or maintain existing ones. For example when reading the following code a new (or existing) contributor may begin to scratch their head 😃
The purpose of this issue is as a place to discuss whether it would be possible to simplify/replace any aspects of the code in relation to the async usage.