September 2019 - Libraries

Strict getContents
by Li-yao Xia 13 Sep '19

13 Sep '19

Hello Libraries, In base, the functions which read all contents from a handle or file into one String currently all do lazy IO: readFile, getContents, hGetContents. https://hackage.haskell.org/package/base-4.12.0.0/docs/System-IO.html#v:hGe… The easiest way to get a strict alternative seems to be to explicitly force the list, for example using ```length contents `seq` pure ()```, but that's far from an obvious solution. Is there a better way? If not, I propose to add readFile', getContents', hGetContents', which don't do lazy IO. It regularly creates confusion among beginners, and it's easy to assume that lazy IO is benign if that's the only way to do certain operations, when it's arguably the wrong way to read files to begin with. Cheers, Li-yao

6 7

Proposal: Add `Generically` (or `WrappedGeneric`) newtype to GHC.Generics
by Alexis King 12 Sep '19

12 Sep '19

Hi all, In recent years, the DefaultSignatures extension has seen popular use as a mechanism for providing derived typeclass instances via GHC.Generics. Although undeniably useful, I have always felt it is somewhat ugly: it is mutually exclusive with other, non-Generic default method implementations, and it can only be used for one Generic deriving mechanism, so implementations must choose between GHC.Generics and Data.Data. Fortunately, with the advent of DerivingVia, there is a better way: simply attach generic instances to a separate newtype, defined like newtype Generically a = Generically { unGenerically :: a } instance Generic a => C (Generically a) where ... then derive instances using DerivingVia as follows: data Foo = Bar X | Baz Y Z deriving C via Generically Foo The `Generically` name already exists for this purpose in the `generic-data` package, making it a good candidate name for a newtype in GHC.Generics (`generic-data` could simply re-export the type with suitably recent versions of `base`). An alternate name would be the more traditional `WrappedGeneric`. I don’t have much of a preference either way, but I do think the `Generically` name is cute, especially when used with DerivingVia. I believe this type should be in `base` because it is (1) clearly generally useful in the same way that similar newtypes in `base` like `WrappedMonad` are (and probably even more so), (2) extremely lightweight in terms of additional API complexity (it’s just a newtype), (3) isn’t worth depending on a separate package for, encouraging a proliferation of (possibly name-conflicting) newtypes in individual packages if it isn’t in `base`, and (4) is an opportunity to add instances based on `Generic` for classes already in `base`. Overall, it’s something that would feel right at home in GHC.Generics to me. As a final note, whichever name people prefer, it would of course make sense to provide an analogous `Generically1` or `WrappedGeneric1` type for `Generic1` (as `generic-data` does as well). Alexis

7 17

Fwd: Proposal: Add singleton function to Data.List module
by Sven Panne 10 Sep '19

10 Sep '19

Am Mi., 21. Aug. 2019 um 12:22 Uhr schrieb Nathan Bouscal < nbouscal(a)gmail.com>: > I would strongly encourage anyone inclined to say things like “there's no benefit > in having the function” to consider reframing the sentiment as “I wouldn’t > benefit from having the function, and don’t understand the benefit others > will gain.” Once you’ve done that reframe, you can notice that there’s a > lack of understanding toward which the appropriate emotional stance is > curiosity, not dismissiveness. A lot of people have clearly expressed in > this thread that they would benefit from this function, and when you assert > as fact that such benefit does not exist, you’re implicitly dismissing and > devaluing their experience. [...] > There is also another side of the coin: A lot of people have clearly expressed counterarguments, so *their* experience has been equally "dismissed" or "devalued". Reframing the discussion shouldn't be biased, otherwise it's just a cheap trick to make one side of the discussion more valuable/stronger. Furthermore, a technical discussion doesn't need to be an excercise in nonviolent communication, as long as people don't start insulting each other. I think the main problem of this discussion (and other similar threads on this list) is that there is no consensus about how to design an API. IMHO, piling up little helpers just because they came up in a few places is a recipe for horrible, bloated APIs. The questions one should ask are "Does it generalize things?", "Does it make my API more powerful?" (i.e. can I do things I couldn't do before), "Does it remove common pitfalls?" etc. But this is just my personal view, obviously others seem to disagree.... :-/

10 13

Re: Proposal: Add singleton function to Data.List module
by Ryan Scott 08 Sep '19

08 Sep '19

> Meanwhile, Ryan GL Scott managed to > accumulate enough data to show that breakage this one combinator would > introduce was pretty minor after all, so we're going to go ahead with > adding singleton before the rest of this happens. In case you're curious what this data looks like: I compiled GHC 8.6.5, but with the singleton function added to Data.List, and built a decently large variety of libraries from Hackage to see what would break. When I was done with this informal experiment, my ~/.cabal/store directory for this version of GHC had 523 entries, which provides a loose upper bound on the number of libraries that successfully built (there are some duplicate entries). You can view these store entries at [1]. Of these ~500 or so libraries, there were a grant total of three that did not build out of the box: * blank-canvas-0.7 * ghc-lib-parser-8.8.0.20190723 * pandoc-2.7.3 Of course, this experiment is not exhaustive—there are many other libraries out there, and there is also lots of code that is not on Hackage. Nevertheless, the overall percentage of libraries that broke was surprisingly small, and certainly much smaller than my initial estimate. Ryan S. ---- [1] https://gitlab.haskell.org/snippets/1474

1 0

optional definition
by Dannyu NDos 07 Sep '19

07 Sep '19

The current 'one or none' definition breaks the order of elements. It is more Ord-friendly to define it as 'none or one'.

4 4