Between the continued iteration on the GHC20xx meta-extension mechanisms, further improvements in the JS/wasm backends, and (my favorite) the availability of exception backtraces in base, there is lots in this release to be excited about.
See the Haskell Discourse thread for the full announcement:
On Wednesday, 2024-05-15, at 1830 UTC, we'll stream the 25th episode of the #Haskell#Unfolder. This episode should be interesting for Haskellers and non-Haskellers alike. Edsko and I will translate a #gRPC server from #Java to Haskell, contrasting the programming paradigms.
Hey, #Haskell, what's the more conservative extension to the report?
I think I can do what I want either way, but I don't really know how to do it in Haskell2010 with no extensions. (I have a non-parameteric type function/family in my semantics, and don't want to give up totality.)
@BoydStephenSmithJr I voted RankNTypes. Arguments in favour: RankNTypes (or at least Rank2Types) existed long before GADTs, and RankNTypes can be translated into "plain" Fω Core, without needing any equality constraints like FC.
Last week I prototyped a git remote helper in a shell script, and now I'm rewriting that in #haskell as part of #gitAnnex.
I don't do this often and I wonder if it was a mistake, probably I should have written the prototype in haskell and then integrated it into git-annex. It's kind of amazing how a lot of complexity is melting away and also how I'm adding So Many Types and also throwing in a lot of robustness improvements.
The #Haskell#Unfolder is now 1 year old! In the anniversary episode, Edsko and I will return to the very first topic and consider (un)folds, but now in a more generic setting. Live on YouTube 2024-05-01 at 1830 UTC.
Brisbane Functional Programming Group #BFPG May meetup, Tue 14th: #Gleam v1 (Rob Ellen) + Do your taxes with #Haskell (Fraser Tweedale). All curious minds are welcome!
#Haskell, #Jones, #Texas National Weather Service: #TORNADO WARNING in this area until 11:15 PM CDT. Take shelter now in a basement or an interior room on the lowest floor of a sturdy building. If you are outdoors, in a mobile home, or in a vehicle, move to the closest substantial shelter and protect yourself from flying debris. Check media. Source: NWS San Angelo TX
AFAIK, there's not an easy way in Haskell to inspect at the type level what type a field has in a record.
What I mean is that that there doesn't seem to be a type family like
type FieldType :: Type -> Symbol -> Type
that we could invoke in ghci like
:kind! FieldType Person "age"
Why would I want this? For libraries like servant and rel8 that use parameterized records where the types of the fields vary heavily with the type parameter.
quoting @prophet
mli files are mostly used to constrain the visibility of definitions whereas hs-boot files are about allowing mutual recursion between modules (which OCaml doesn't support, even with mli files!)
But the mechanism by which they achieve their goals is nearly identical even though the perception of it is so vastly different.
I guess the conclusion to draw from this is that both sides are wrong: IMO, mli files are not nearly as good as OCamlers think they are, but hs-boot files aren't as ugly as Haskellers think either.
-- prettySrcLoc and prettyCallStack are defined here to avoid hs-boot
-- files. See Note [Definition of CallStack]
Backpack's design is primarily driven by compatibility considerations (“how do we build upon GHC's existing foundation?”), rather than elegance. In particular, Backpack doesn't eliminate those ugly .hs-boot files, it just automates and hides their generation and processing.
In Servant, the ServerError type has an Exception instance https://hackage.haskell.org/package/servant-server-0.20/docs/Servant-Server.html#t:ServerError
You might speculate that when throwing a ServerError using liftIO . throwIO in a Handler, the ServerError is automatically caught and served as a response, but it ain't so: it's treated as just another exception, and the response code is 500.
