Clojurians Log v2

niwinz: i see there's a load of new stuff in cats... will there be a release soon ?

We have not yet thought about it

I think that the main blocking part is traversable

and applicative composition

i don't know if we should leave the last (applicative composition) for other release or include it in this one.

by applicative composition dyu mean the alet macro ?

nope, I mean compose deferred with either

and use it in alet

basically extend the transformers to support applicatives

ah, yeah, i remember

In any case the reall blocking issue is a traversable

what is traversable ?

https://hackage.haskell.org/package/base-4.8.1.0/docs/Data-Traversable.html

This link maybe will explain better that I can

The next cats release will be the first 1.0.0 i think

so you are adding a new protocol for traversable data types ?

Yes https://github.com/funcool/cats/pull/80

@dialelo: is working on it

yes

the idea behind traverse is to

map every element to an acction (applicative)

and evaluate such actions from left to right using applicative bind

a bit more general than map and reduce, you can both map & combine in single traversal

i hope to have this ready by the weekend

i travel on friday and will use that time for finishing this PR

cool

dyu have any plans to add comonads and arrows to cats ?

i don't have any particular need for them atm, just wondering

we could, of course

i have an intuition of what Comonads look like being the dual of monads

but i'd have to learn more about the subject

same with arrows

It seems that is a great task for new contributors 😛

mccraigmccraig: feel free to open issues in the cats repo, we use github to track remaining work on the library

niwinz: :simple_smile: when i get a bit of relief from the delivery schedule i'm on atm i would love to work on some of this stuff - understanding and using it is certainly improving my programming

\o/

indeed!

oh, i had a question about cats.labs.deferred ... why do -pure/-return use a manifold/future rather than a simple manifold/success-deferred ?

Hmm, because I don't know about success-deferred

didn't know

ha, ok - future uses another thread, so i was wondering if there was some deeper purpose i hadn't grokked :simple_smile:

No, knowing about success-deferred, I will change the pure/return impl to use it.

Additional thread is useless in this case.