- Haskell-Cafe - Haskell.org

Re: [Haskell-cafe] Haskell-Cafe Digest, Vol 207, Issue 16
by Sumit Raja 19 Nov '20

19 Nov '20

> It may be that you have a dependency of this specific test (or the test > itself) that has the following stanza in its Cabal file: > "extra-libraries: gmp" > > Thanks, makes sense. Any shortcuts to try and find which package has gmp enabled? Interestingly this problem doesn't happen with ghc-8.6.5. Does build order decide on if this issue is triggered or not? -Sumit

2 1

ghci module re exporting qualified as
by Immanuel Litzroth 18 Nov '20

18 Nov '20

I'm trying to make a test module to run tests. I want to import the test functionality + imported functionality in the scope of the test module under ghci. I also want to have short names in the Test module for reexported modules. I am using Data.Map and Data.Set ... so lot's of conflicting exports. Is this possible? #+BEGIN_SRC haskell :tangle T1.hs module T1 ( module Data.Map) where import qualified Data.Map #+END_SRC works but no browsing info shows up #+BEGIN_EXAMPLE λ⊢ :show modules T1 .... λ⊢ :show imports import T1 -- added automatically import Prelude -- implicit λ⊢ fromList [] <interactive>:114:1-8: error: Variable not in scope: fromList ∷ [a0] → t λ⊢ Data.Map.fromList [] fromList [] λ⊢ :browse! T1 λ⊢ #+END_EXAMPLE #+BEGIN_SRC haskell :tangle T2.hs module T2 ( module Map) where import qualified Data.Map as Map #+END_SRC #+BEGIN_EXAMPLE λ⊢ :show modules T2 :show imports import T2 -- added automatically import Prelude -- implicit λ⊢ fromList [] <interactive>:119:1-8: error: Variable not in scope: fromList ∷ [a0] → t λ⊢ Map.fromList [] <interactive>:120:1-12: error: Not in scope: ‘Map.fromList’ No module named ‘Map’ is imported. λ⊢ Data.Map.fromList [] fromList [] λ⊢ :browse! T2 λ⊢ #+END_EXAMPLE -- -- Researching the dual problem of finding the function that has a given point as fixpoint.

3 5

Why is my 'text' missing an instance?
by amindfv＠mailbox.org 18 Nov '20

18 Nov '20

tl;dr: I seem to have two versions of 'text' on my system, both of which should have an instance for (Binary Text). However, one version seems to claim not to have that instance. $ cat test.hs import qualified Data.Text as T import qualified Data.Binary as B main = print $ B.encode $ T.pack "hello" $ /usr/local/bin/ghci --version The Glorious Glasgow Haskell Compilation System, version 8.10.2 $ /usr/local/bin/ghci test.hs GHCi, version 8.10.2: https://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( test.hs, interpreted ) test.hs:4:16: error: • No instance for (B.Binary T.Text) arising from a use of ‘B.encode’ • In the second argument of ‘($)’, namely ‘B.encode $ T.pack "hello"’ In the expression: print $ B.encode $ T.pack "hello" In an equation for ‘main’: main = print $ B.encode $ T.pack "hello" | 4 | main = print $ B.encode $ T.pack "hello" | ^^^^^^^^^^^^^^^^^^^^^^^^^ Failed, no modules loaded. $ /usr/local/bin/ghc-pkg list text /usr/local/lib/ghc-8.10.2/package.conf.d text-1.2.3.2 /home/name/.ghc/x86_64-linux-8.10.2/package.conf.d text-1.2.4.1 $ ghc-pkg-8.10.2 list text [same result] $ /usr/local/bin/ghci -package text-1.2.3.2 test.hs GHCi, version 8.10.2: https://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( test.hs, interpreted ) Ok, one module loaded. $ /usr/local/bin/ghci -package text-1.2.4.1 test.hs GHCi, version 8.10.2: https://www.haskell.org/ghc/ :? for help [1 of 1] Compiling Main ( test.hs, interpreted ) test.hs:4:16: error: • No instance for (B.Binary T.Text) arising from a use of ‘B.encode’ • In the second argument of ‘($)’, namely ‘B.encode $ T.pack "hello"’ In the expression: print $ B.encode $ T.pack "hello" In an equation for ‘main’: main = print $ B.encode $ T.pack "hello" | 4 | main = print $ B.encode $ T.pack "hello" | ^^^^^^^^^^^^^^^^^^^^^^^^^ Failed, no modules loaded. Both text-1.2.3.2 and text-1.2.4.1 have a Binary instance for Text. What's going on? (After writing this out I rebuilt an unmodified clone of text-1.2.4.1 but I see the same results.) Thanks, Tom

4 4

GHC linker error 8.8.4 Windows
by Sumit Raja 17 Nov '20

17 Nov '20

Hi, Suddenly started getting the following linker error after I moved code around in my project. It only seems to occur when I am building one specific cabal test target. When building a dependent build it build successfully but cabal repl test-xxx fails. I have no consistent way to repro but I will keep trying and file a defect.. GHCi, version 8.8.4: https://www.haskell.org/ghc/ :? for help GHC runtime linker: fatal error: I found a duplicate definition for symbol __gmpn_sec_sqr_itch whilst processing object file C:\Users\sumit\Apps\ghc-8.8.4\mingw\lib\libgmp.dll.a The symbol was previously defined in C:\Users\sumit\Apps\ghc-8.8.4\lib\integer-gmp-1.0.2.0\HSinteger-gmp-1.0.2.0.o This could be caused by: * Loading two different object files which export the same symbol * Specifying the same object file twice on the GHCi command line * An incorrect `package.conf' entry, causing some object to be loaded twice. ghc.exe: panic! (the 'impossible' happened) (GHC version 8.8.4 for x86_64-unknown-mingw32): loadArchive "C:\\Users\\sumit\\Apps\\ghc-8.8.4\\mingw\\lib\\libgmp.dll.a": failed Please report this as a GHC bug: https://www.haskell.org/ghc/reportabug CallStack (from HasCallStack): die', called at .\\Distribution\\Client\\ProjectOrchestration.hs:1041:55 in main:Distribution.Client.ProjectOrchestration cabal.exe: repl failed for mkr-txn-0.1.0.0-inplace-test-mkr-txn. Regards Sumit

2 1

Re: [Haskell-cafe] Reading Haddock sources on hackage no longer possible with simple browsers?
by M Douglas McIlroy 16 Nov '20

16 Nov '20

> I am pretty sure nobody writes code like this > I confirmed your example, reading with lynx and w3m. > wouldn’t just directly opening the sources (in a text editor) be simpler? Is there a (simple) way to preserve the experience when browsing using Lynx without also holding back features aimed at a regular browsing experience? I second that. I have always been annoyed by the ransom-note appearance of colorized listings. But that's just my taste. I suspect, though, that lots of people have been frustrated when they discover they can't use the code they see because it's festooned with markup. Golang.org offers similar listings, but also provides a button for selecting the underlying plain text. I would like hackage a lot better if it did so, too. Doug

3 2

[ANN] base64-bytestring-1.2.0.1 + future plans
by Emily Pillmore 15 Nov '20

15 Nov '20

Hi Everyone, I'm pleased to announce that we've released a new version of `base64-bytestring` that now supports `bytestring-0.11` in a backwards-compatible manner. I'd also like to make an RFC: we are thinking that we will drop support for GHC versions <7.10 in future releases of the project. The earliest release in which this can be expected will be around 6 months from now. The dev burden for supporting back to 7.0 is high, and unless someone screams in protest, the value does not seem to be there. I'd like to get your thoughts on this. Cheers, Emily

1 0

Redirecting hackage.haskell.org to HTTPS
by Tikhon Jelvis 15 Nov '20

15 Nov '20

Pages on hackage.haskell.org are currently served over both HTTP and HTTPS. Package security on Hackage does not depend on HTTPS, so we keep HTTP endpoints for backwards compatibility with automated systems that depend on Hackage and do not support HTTPS. However, this means that it's possible for users to inadvertently browse Hackage pages on HTTP which is not a great user experience. To address this issue without breaking existing scripts, we are planning to redirect requests to HTTPS based on User-Agent headers: requests with "Mozilla/5.0" in their User-Agent string will be redirected to HTTPS and other requests will remain unchanged. Please contact us at committee(a)haskell.org if this change will cause problems with how you use Hackage. Otherwise, the new behavior will go into effect on 2020-11-23. Thanks! -Tikhon Jelvis, on behalf of the Haskell.org Committee

3 2

Re: [Haskell-cafe] Reading Haddock sources on hackage no longer possible with simple browsers?
by Johannes Waldmann 15 Nov '20

15 Nov '20

> I recently noticed that it is no longer possible > to read the source of a module, because extra type information > (which is likely hidden in a modern browser) > is inline with the actual sources. I confirmed your example, reading with lynx and w3m. It seems the extra text is the methods of the type classes (mentioned after "deriving") and it gets shown on on mouse-over in a graphical browser. (Re: "nobody writes code like this" - the compiler does? Some methods here seem auto-generated.) Looks like a bug in haddock, then? (Or in lynx' handling of CSS?) The actual HTML source looks legit, at least partially (see example below) This seems related: https://github.com/haskell/haddock/pull/1197 I will open an issue on accessibility. - J.W. deriving (Int -> ISBN -> ShowS [ISBN] -> ShowS ISBN -> String (Int -> ISBN -> ShowS) -> (ISBN -> String) -> ([ISBN] -> ShowS) -> Show ISBN forall a. (Int -> a -> ShowS) -> (a -> String) -> ([a] -> ShowS) -> Show a showList :: [ISBN] -> ShowS $cshowList :: [ISBN] -> ShowS show :: ISBN -> String $cshow :: ISBN -> String showsPrec :: Int -> ISBN -> ShowS $cshowsPrec :: Int -> ISBN -> ShowS Show

4 4

Reading Haddock sources on hackage no longer possible with simple browsers?
by Mario Lang 14 Nov '20

14 Nov '20

Hi. I have been using Lynx to read stuff on hacakge for a long time. I recently noticed that it is no longer possible to read the source of a module, because extra type information (which is likely hidden in a modern browser) is inline with the actual sources. This makes the go to source feature effectively useless. An example (from the ISBN package): deriving (Int -> ISBN -> ShowS [ISBN] -> ShowS ISBN -> String (Int -> ISBN -> ShowS) -> (ISBN -> String) -> ([ISBN] -> ShowS) -> Show ISBN forall a. (Int -> a -> ShowS) -> (a -> String) -> ([a] -> ShowS) -> Show a showList :: [ISBN] -> ShowS $cshowList :: [ISBN] -> ShowS show :: ISBN -> String $cshow :: ISBN -> String showsPrec :: Int -> ISBN -> ShowS $cshowsPrec :: Int -> ISBN -> ShowS Show, ISBN -> ISBN -> Bool (ISBN -> ISBN -> Bool) -> (ISBN -> ISBN -> Bool) -> Eq ISBN forall a. (a -> a -> Bool) -> (a -> a -> Bool) -> Eq a /= :: ISBN -> ISBN -> Bool $c/= :: ISBN -> ISBN -> Bool == :: ISBN -> ISBN -> Bool $c== :: ISBN -> ISBN -> Bool Eq) I know Haskell has a ton of extensions, but I am pretty sure nobody writes code like this :-) This is, frankly, pretty horrible. Can something be done to fix this? (I am aware that certain modernisations are unstoppable. However, this one looks like a bug.) -- CYa, ⡍⠁⠗⠊⠕

2 1

Generic Sorting
by Henry Laxen 14 Nov '20

14 Nov '20

Good evening Haskellers. I've done some searching but so far haven't found anything, which make me think this probably isn't possible. I am wondering if it is possible to do a "Generic Sort" on multilevel data structures. Suppose you have something like: data A = A Int [Int] data B = B Int [A] a1 = A 2 [2,1] a2 = A 1 [4,3] b = B 1 [a1,a2] genericSort b = B 1 [A 1 [3,4], A 2 [1,2]] note that not only are the A's sorted, but the list inside each A is sorted. It "sorted all the way down". Has Edward written such a thing yet? ;-) His "discrimination" library doesn't do this, in case you're wondering. Any ideas? Best wishes, Henry Laxen

3 3