Levi Ramsey

It's in the Scala portion of the codebase: https://github.com/RWS/odata/blob/500cdb6dfa69197fb135867a7425cd3c85682bf9/odata_api/src/main/scala/com/sdl/odata/api/processor/query/ODataQuery.scala#L31

I believe the extra `( )` around function parameters is a Scala 3 requirement (at least I've regularly had to clean up my code with parens around single arguments to...

Scala 3 requires parens around single arguments with type ascriptions. Perhaps scalafmt also imposes the parens for its Scala 3-compatible modes?

As noticed in another repo, `docstrings.wrap = yes` (the default) will do weird things with `@param` etc. in scaladoc, so incorporating that here.

This operator doesn't seem to make sense for the unending stream use case (universal quantification and all that), where a `Sink.exists` though would make sense (e.g. to build a "this...

`Sink.exists(predicate)` could be approximately implemented, though, as `Flow[T].filter(predicate).toMat(Sink.head) { (_, headFut) => headFut.map(_ => true).recoverWith { case _: NoSuchElementException => Future.successful(false) }`...

Can we be sure that we're not making things worse by swallowing the stack overflow? Consider that the stack overflow could happen when an invariant of some object graph is...

If `BlockingQueue` is the reason, then that would argue for using `whenCompleteAsync` on the `CompletionStage` variant (at least if the EC is an `ExecutionContextExecutor`), would it not?

One would idiomatically implement both in Akka using actors (and event sourcing). It's a somewhat different model from those here, which is part of why it might be interesting.

I suspect this is due to the `Create` tasks being incremental. An `sbt clean` before `headerCreateAll` seems to address this for me.