Gabriel Scherer
Gabriel Scherer
(My comment here had been placed in the wrong issue, apologies and thanks @dbuenzli for the notification.)
The code looks fine to me. The question is whether the benefits of fixing bad behaviors on 4.14 is worth the risk of fixing accidentally-desirable behaviors in 4.14. I think...
As I stated during the meeting, I believe that we should give ourselves a chance to design a static type/effect system for effects, and I hope to get a better...
If you are happy to write everything in an environment monad, then you can also abstract over the function `(fun v -> perform (Yield v))`, and thus add a `yield:('a...
> I'm most worried by the added complexity to the type-checker if upstreaming this as a configure switch, as the `Tstr_effect` changes would be present in the Typedtree, but never...
A note for myself: I think that the `continue with ` syntax comes from the internal implementation approach for shallow handlers in Multicore OCaml, which are implemented exactly like deep...
I have a slightly less grumpy take on this feature request than @Octachron, a nice change of role as I'm usually the bad cop. I think that, yes, it would...
Interesting, thanks! I think that I disagree with some of the category names. For example "The method `x#m`" reads wrong to me, `x#m` is a method call, `m` is the...
Do we want to revisit this or close it?
I tend to agree with Daniel: I prefer a world where environment variables used to determine the behavior of an OCaml program are read at an unspecified time during the...