Lack of whileM, whileM_, untilM and untilM_ methods for Monad

[Atry]

These methods present in scalaz.Monad. Should we add them in cats as well?

[kailuowang]

Forgive my ignorance ( I never used these before), I am just curious is there any usecases for this other than the MonadState, in particular, MonadState[StateT[Trampoline, S, ?], S (with Stack safety) or MonadState[State[S, ?], S] (without) .

[peterneyens]

There is this closed PR #1216 by @tpolecat (from before the move of tailRecM to FlatMap).

[tpolecat]

@kailuowang these can be useful for IO-like types, or things that will be interpreted into IO … in doobie for instance get[Foo].whileM[Vector](next) is a program of type Free[ResultSetOp, Vector[Foo]] that expresses the intent to read a resultset into a vector.

I closed that PR because I lost interest in the tailRecM discussion, but it should be straightforward to implement now.

[Atry]

@kailuowang I am porting ThoughtWorks Each to cats.

At the moment, ThoughtWorks Each requires whileM_ on scalaz.Monad in order to transform do and while loops.

I will be able to transform do and while loops on cats, if cats support whileM_ , too.

[johnynek]

Now that we have tailRecM on FlatMap these methods are easy (and stack safe).

…

On Wed, Mar 22, 2017 at 15:33 杨博 (Yang Bo) ***@***.***> wrote: @kailuowang <https://github.com/kailuowang> I am porting ThoughtWorks Each <https://github.com/ThoughtWorksInc/each> to cats. At the moment, ThoughtWorks Each <https://github.com/ThoughtWorksInc/each> requires whileM_ on scalaz.Monad in order to transform do and while loops. I will be able to transform do and while loops on cats, if cats support whileM_, too. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1569 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAEJduBZwMQ8nCBaumb3BEiul24GoGNBks5rocvpgaJpZM4Mk-L_> .