fix: Added priority queue sink and source extending queue

[neomaclin]

trying to continue the work #3733 by @lukecollier
@armanbilge

[durban]

It seems to me, that tryTakeN and tryOfferN has the same implementation in QueueSource/QueueSink. So why are we overriding?

[neomaclin]

It seems to me, that tryTakeN and tryOfferN has the same implementation in QueueSource/QueueSink. So why are we overriding?

actually, come to think of it. they are actually identical in almost everyway. a simple extends without should just do., pushed

[neomaclin]

insteresting seeing the error in Mima check, maybe that is need of override?

[armanbilge]

insteresting seeing the error in Mima check, maybe that is need of override?

yes, I think you can add an override call that delegates to the super so you do not need to repeat the implementation.

[neomaclin]

insteresting seeing the error in Mima check, maybe that is need of override?

yes, I think you can add an override call that delegates to the super so you do not need to repeat the implementation.

looks like Mima hates either way....(override with super call triggered the ReversedMissingMethodProblem)

[neomaclin]

why DispatcherSpec failed is a bit beyound me...

[neomaclin]

It seems to me, that tryTakeN and tryOfferN has the same implementation in QueueSource/QueueSink. So why are we overriding?

I guess the attempts were telling us MiMa doesn't agree...

[armanbilge]

looks like Mima hates either way....(override with super call triggered the ReversedMissingMethodProblem)

hmm. can you share the full MiMa error message? I don't understand why it's not happy.

[neomaclin]

looks like Mima hates either way....(override with super call triggered the ReversedMissingMethodProblem)

hmm. can you share the full MiMa error message? I don't understand why it's not happy.

ah. in order to fix the build, I have force pushed and squashed that commit (not sure how to find the actual error log in that case). but bascially MiMa is reported the "ReversedMissingMethodProblem" along with the message "super.tryOfferN is present only in the current version" if I have only

override def tryOfferN(list: List[A])(implicit F: Monad[F]): F[List[A]] = super.tryOfferN(list)

[neomaclin]

looks like Mima hates either way....(override with super call triggered the ReversedMissingMethodProblem)

hmm. can you share the full MiMa error message? I don't understand why it's not happy.

or, I could break the build again to generate the error.

[neomaclin]

looks like Mima hates either way....(override with super call triggered the ReversedMissingMethodProblem)

hmm. can you share the full MiMa error message? I don't understand why it's not happy.

here they are

[error] cats-effect-std: Failed binary compatibility check against org.typelevel:cats-effect-std_2.13:3.3.11 (e:info.apiURL=https://typelevel.org/cats-effect/api/3.x/, e:info.versionScheme=early-semver)! Found 2 potential problems (filtered 18)
[error]  * abstract synthetic method cats$effect$std$PQueueSink$$super$tryOfferN(scala.collection.immutable.List,cats.Monad)java.lang.Object in interface cats.effect.std.PQueueSink is present only in current version
[error]    filter with: ProblemFilters.exclude[ReversedMissingMethodProblem]("cats.effect.std.PQueueSink.cats$effect$std$PQueueSink$$super$tryOfferN")
[error]  * abstract synthetic method cats$effect$std$PQueueSource$$super$tryTakeN(scala.Option,cats.Monad)java.lang.Object in interface cats.effect.std.PQueueSource is present only in current version
[error]    filter with: ProblemFilters.exclude[ReversedMissingMethodProblem]("cats.effect.std.PQueueSource.cats$effect$std$PQueueSource$$super$tryTakeN")

[neomaclin]

fb356b7 with the implementaion on the overrided method was ok with MiMa (and other build checks). let me know if I should go back to that for succcess build of the PR. oh. and since I don't have the write access, I wouldn't be able to merge anything even if the build is/was ok.

[armanbilge]

We should avoid duplicating the implementation. Since the super trick is not possible, we can extract the implementation to a method on the companion object, and then both traits can delegate to that.

[neomaclin]

some how the DispatcherSpec comes back

[error]     x raise an error on leaked runner
[error]      Expected: java.lang.IllegalStateException. Got nothing (DispatcherSpec.scala:269)

[samspills]

some how the DispatcherSpec comes back

[error]     x raise an error on leaked runner
[error]      Expected: java.lang.IllegalStateException. Got nothing (DispatcherSpec.scala:269)

@neomaclin that's unrelated to your PR, don't worry :) It's a known flake: #3776

[neomaclin]

if this is good to go in, please go ahead, ( I can't do it myself..)

[djspiewak]

Thank you for taking this on!