December 2015 - Haskell-Cafe

Hyperlinked source docs
by Johannes Waldmann 16 Dec '15

16 Dec '15

As I said, I admire hyperlinked sources in the new "base" documentation https://hackage.haskell.org/package/base-4.8.1.0/docs/ and I want to know how I can produce such documents. What is the magic haddock/hscolour invocation? (I actually use standalone-haddock.) On hackage, not every package gets the special treatment? (In fact, only "base" does?) - J.W.

5 4

IHaskell trouble, after upgrading to Haskell Platform 7.10.3?
by David Banas 16 Dec '15

16 Dec '15

Hi all, Are there any other IHaskell users out there, who are having trouble re-installing IHaskell, after upgrading to Haskell Platform 7.10.3? I’m getting this, at the tail end of the actual ihaskell build: [24 of 25] Compiling IHaskell.Eval.Info ( src/IHaskell/Eval/Info.hs, dist/build/IHaskell/Eval/Info.o ) [25 of 25] Compiling IHaskell.Publish ( src/IHaskell/Publish.hs, dist/build/IHaskell/Publish.o ) In-place registering ihaskell-0.8.3.0... Preprocessing executable 'ihaskell' for ihaskell-0.8.3.0... main/Main.hs:29:18: Could not find module ‘Data.String.Here’ It is a member of the hidden package ‘here-1.2.7@here_63mTmLx8Sz20TyEJBVCnnC’. Perhaps you need to add ‘here’ to the build-depends in your .cabal file. Use -v to see a list of the files searched for. Updating documentation index /Users/dbanas/Library/Haskell/share/doc/x86_64-osx-ghc-7.10.3/index.html cabal: Error: some packages failed to install: ihaskell-0.8.3.0 failed during the building phase. The exception was: ExitFailure 1 Thanks, and Happy Holidays! -db

2 2

Parallel Haskell on RPi2
by spacestation＠venussociety.org 16 Dec '15

16 Dec '15

Hello to everone, Did anybody manage to install parconc-examples on a Raspberry Pi2 without the use of --force-reinstalls. As it would break to many packages on my end.I tried it the way describe from Simon Marlow s book "Parallel and concurrent programming in haskell". Or is there any option within cabal to seperate between new,new version and reinstall packages listed with pi@opencvrpi2:~/parconc-examples-master$ cabal install --only-dependencies I m using ghc 7.4.1 Thanks for any feedback Gottfried F. Zojer

1 0

sum $ map f xs ... ghc-7.10 performance regression?
by Johannes Waldmann 15 Dec '15

15 Dec '15

Is there a difference between "sum $ map f xs" and "sum $! map f xs"? I would think no, since "$!" just evaluates the argument to WHNF, so it would evaluate just the first cons of the list. This little program main = print $ sum $ map bitcount [0, 4 .. 2^24-1 ] bitcount :: Int -> Int bitcount x = if x > 0 then let (d,m) = divMod x 2 in bitcount d + m else 0 runs in 1.6 seconds when compiled with -O2 on ghc-7.6.3, ghc-7.8.4, but takes 3.2 seconds on ghc-7.10.[1-3]. Why? when I change the main function (note: $ => $!) to main = print $ sum $! map bitcount [0, 4 .. 2^24-1 ] and compile with 7.10.[1-3], then it also runs in 1.6 seconds. (I can write a bug report but I want to check whether I am missing something obvious) - J.W.

5 8

Does Backwards admit a Monad instance?
by David Feuer 15 Dec '15

15 Dec '15

In particular, I'm thinking about instance MonadFix m => Monad (Backwards m) where m >>= f = Backwards $ do fin <- forwards (f int) int <- forwards m return fin It looks to me like this should be valid, and compatible with the Applicative instance, but maybe I'm missing something.

1 1

Re: [Haskell-cafe] Trouble installing ghc on CentOS 6
by Johannes Waldmann 15 Dec '15

15 Dec '15

I think that "--with-gmp-libraries" states the location of libgmp that the installed ghc will use when linking programs. Your error indicates that a .so file referenced from the executable (contained in the binary package) "utils/ghc-pwd/dist-install/build/tmp/ghc-pwd" could not be resolved. $ ldd utils/ghc-pwd/dist-install/build/tmp/ghc-pwd if it says ... libgmp.so.3 => not found then set LD_LIBRARY_PATH or check values in /etc/ld.so.conf and run /sbin/ldconfig - J.W.

2 1

Trouble installing ghc on CentOS 6
by Lyle Kopnicky 14 Dec '15

14 Dec '15

Hi folks, I'm trying to install the 64-bit generic Linux distribution of ghc-7.10.3 on CentOS 6, as a non-privileged user in my home dir. I've unpacked the files, and I ran ./configure, but it told me it couldn't find libgmp.so.3. I searched for and found that library on the system. Looking at the help for ./configure, I saw the --with-gmp-libraries option, so I tried again with that, but it still can't seem to find the file: $ ./configure --prefix=/home/kopnicky/ghc --with-gmp-libraries=/usr/lib64 checking for path to top of build tree... utils/ghc-pwd/dist-install/build/tmp/ghc-pwd: error while loading shared libraries: libgmp.so.3: cannot open shared object file: No such file or directory configure: error: cannot determine current directory Strange, because that library is definitely there in /usr/lib64. Any thoughts on what I can try next? Thanks, Lyle

1 0

a problem I can't solve simply
by Dennis Raddle 14 Dec '15

14 Dec '15

I have a problem that doesn't seem hard to state but I can't seem to solve without a bunch of complex code. This relates to my musical score playback. In using it to play music, I don't always want to play back the entire source musical document, but rather play a range of measures. So I might give a command to my app like "play 1-3" which means play measures 1 through 3. There is a time saving feature, which is that I can type "play 10" which means start the playback at measure 10 and continue until the first occurrence of two empty measures. This is a common use case. So I have to write a function that takes a start measure and computes the end measure by scanning for two empty measures. Let's say for simplicity's sake that we'll forget about "measures" and just say that notes have a start time and end time, which will be integers. type Note = (Int,Int) A musical score can have several individual staves (notes for individual instruments), so it will look like this: type Staff = [Note] type Score = [Staff] I need to write a function as follows computeEndMsr :: Int -> Score -> Int computeEndMsr beginMsr score = ... Some examples: Here's a score with just one staff, to give you an idea. score1 = [ [(1,3), (2,4), (7,10)] ] -- In the following case a two-unit gap is found at units 5 and 6. computeEndMsr 1 score1 = 4 computeEndMsr 5 score1 = should throw an error indicating that a gap was found immediately and no actual notes were included -- In the following case, the maximum unit of any note is 10, so that is what is computed computeEndMsr 6 score1 = 10 -- This case illustrates how it's okay if the computed end measure is equal to the begin msr computeEndMsr 10 score1 = 10 computeEndMsr 11 score1 = should throw an error indicating that the given begin msr is past the end of any note in the score This example has only one staff, but a score can have multiple staves. Also the timing and duration of notes can overlap, either on one staff or across multiple staves.

2 2

Applicative banana brackets
by martin 14 Dec '15

14 Dec '15

Hi, while learning about all the type classes and their relationships I came across something I found weird. If I understand it correctly, banana brackets where originally developed for Applicatives. The intent was to enable us to write something like (| (\a b c -> a + b + c), [3], [1,2], [5,0,7] |) and have it translated to liftA3 (\a b c -> a + b + c) [3] [1,2] [5,0,7] or alternatively, to allow us to write something like (| (pure $ \a b c -> a + b + c), [3], [1,2], [5,0,7] |) and have it translated directly to pure (\a b c -> a + b + c) <*> [3] <*> [1,2] <*> [5,0,7] A variant of banana brackets is implemented in ghc, but only for Arrows as part of -XArrowSyntax. Arrows are just the intersection of Applicative and Category, so this implementation seems to be a specialization. What's worse, I don't think -XRebindableSyntax extends to banana brackets. But why? Is it hard to have the notation work with both? (After all, the relationship between Arrows and Applicatives is not easily expressed in Haskell.) Was the demand for (Applicative) bananas not big enough? Is it just a relic? And more to the point: I have not looked at the ghc code base at all yet, but it's on my bucket list to hack on it one day. Right now, a generalization of banana brackets seems like a simple enough low pressure first project, but I fear that it might break code or that there is some pitfall I'm not seeing. Can anybody shed a bit of light on this? Thanks and cheers, Martin L. P.S.: If the list receives this mail several times, I apologize. The list management tool seems to be confused by gmail vs. googlemail. That's what you get for using non-Haskell software. ;)

7 13

[NL-FP 2016] Final CFP: Dutch Functional Programming Day 2016
by Atze Dijkstra 14 Dec '15

14 Dec '15

[My apologies for multiple received copies of the same message] Final Call for Participation: Dear all, The next Dutch Functional Programming day (NL-FP 2016) will be held on Friday, January 8, 2016 at the Utrecht University, The Netherlands. The program of the day now is online (see the below webpage). At the end of the day we will have a joint dinner. Besides the program, on the web page http://foswiki.cs.uu.nl/foswiki/NlFpDay2016/WebHome <http://foswiki.cs.uu.nl/foswiki/NlFpDay2016/WebHome> for the day you’ll also find the registration, which is by letting me know via email (with a subject header that begins with [NL-FP 2016]) whether you want to (1) participate, (2) join dinner, also see the webpage for additional details. There is still room for a demo or poster during lunch/break time. If you intend to participate, please let me know before the end of this year because we like to know how many people we can expect. Hope to meet you all in Utrecht at the next FP Day! Best regards, - Atze - Atze Dijkstra, Department of Information and Computing Sciences. /|\ Utrecht University, PO Box 80089, 3508 TB Utrecht, Netherlands. / | \ Tel.: +31-30-2534118/1454 | WWW : http://www.cs.uu.nl/~atze <http://www.cs.uu.nl/~atze> . /--| \ Fax : +31-30-2513971 .... | Email: atze(a)uu.nl <mailto:atze@uu.nl> ............... / |___\

1 0