February 2015 - Beginners

Re: [Haskell-beginners] tower hanoi problem
by Doug McIlroy 18 Feb '15

18 Feb '15

> My way of working is one problem at the time. > So first solve the itterate one and after that I gonna try to solve the > recursion one. > Otherwise I get confused. This is the crux of the matter. You must strive to think those thoughts in the opposite order. Then you won't get confused. Recursion takes a grand simplifying view: "Are there smaller problems of the same kind, from the solution(s) of which we could build a solution of the problem at hand?" If so, let's just believe we have a solver for the smaller problems and build on them. This is the recursive step. Of course this can't be done when you are faced with the smallest possible problem. Then you have to tell directly how to solve it. This is the base case. [In Hanoi, the base case might be taken as how to move a pile of one disc to another post. Even more simply, it might be how to move a pile of zero discs--perhaps a curious idea, but no more curious than the idea of 0 as a counting number.] A trivial example: how to copy a list (x:xs) of arbitrary length. We could do that if we knew how to copy the smaller list tail, xs. All we have to do is tack x onto the head of the copy. Lo, the recipe copy (x:xs) = x : copy xs Final detail: when the list has no elements, there is no smaller list to copy. We need another rule for this base case. A copy of an empty list is empty: copy [] = [] With those two rules, we're done. No iterative reasoning about copying all the elements of the list. We can see that afterward, but that is not how we got to the solution. [It has been suggested that you can understand recursion thus > Do the first step. Then (to put it very dramatically) > do *everything else* in *a single step*! This point of view works for copy, and more generally for "tail recursion", which is trivially transformable to iteration. It doesn't work for Hanoi, which involves a fancier recursion pattern. The "smaller problem(s)" formulation does work.] In many harder problems a surefire way to get confused is to try to think about the sequence of elementary steps in the final solution. Hanoi is a good case in point. Eventually you will come to see recursion as a way to confidently obtain a solution, even though the progress of the computation is too complicated to visualize. It is not just a succinct way to express iteration! Doug McIlroy

3 3

Re: [Haskell-beginners] Getting "Discrete Mathematics Using a Computer" code to run
by Henk-Jan van Tuyl 15 Feb '15

15 Feb '15

There are several types of exceptions that could be caught here, but the compiler does not know which one to choose. You could specify a type, but it is easier to use 'onException' (also in Control.Exception) instead of 'catch', if you want to treat all exceptions the same way, without any info about the kind of exception. On Sun, 15 Feb 2015 17:49:21 +0100, Lawrence Bottorff <borgauf(a)gmail.com> wrote: > Thanks for the tip. Yes, I had tried that, but never got it right. Here's > my change: : > Stdm02142015.lhs:1037:40-46: > No instance for (Exception e1) arising from a use of `catch' > The type variable `e1' is ambiguous > Possible fix: add a type signature that fixes these type variable(s) > Note: there are several potential instances: > instance Exception NestedAtomically > -- Defined in `Control.Exception.Base' > instance Exception NoMethodError > -- Defined in `Control.Exception.Base' > instance Exception NonTermination > -- Defined in `Control.Exception.Base' > ...plus 7 others > In a stmt of a 'do' block: > filedata <- (readFile filename) `catch` (\ _ -> return "") : > et cetera, et cetera. Any idea where this import Control.Exception should > go. Or is there something else amiss? > > On Sun, Feb 15, 2015 at 5:06 AM, Henk-Jan van Tuyl <hjgtuyl(a)chello.nl> > wrote: > >> On Sat, 14 Feb 2015 23:30:13 +0100, Lawrence Bottorff >> <borgauf(a)gmail.com> >> wrote: >> >> .../Discrete/Stdm02142015.lhs:1036:40-46: >>> Not in scope: `catch' >>> >>> .../Discrete/Stdm02142015.lhs:1048:40-46: >>> Not in scope: `catch' >>> Failed, modules loaded: none. >>> >>> This is on ghci 7.6.3 >>> >> >> The function "catch" is removed from Prelude, you need to add the line: >> import Control.Exception >> >> Regards, >> Henk-Jan van Tuyl -- Folding@home What if you could share your unused computer power to help find a cure? In just 5 minutes you can join the world's biggest networked computer and get us closer sooner. Watch the video. http://folding.stanford.edu/

1 0

Re: [Haskell-beginners] Spec with QuickCheck generationg list of list of the same length
by Ondrej Nekola 15 Feb '15

15 Feb '15

Thanks, it sent me to the (higly probably) right direction: newtype RectList = RectList [LVector] deriving Show instance Arbitrary RectList where arbitrary = do m <- getPositive <$> arbitrary n <- getPositive <$> arbitrary RectList <$> replicateM n (vector m) and the test can look like this: it "average times number is sum" $ property (\(RectList x) -> null $ nozeroMembers $ zipWith (-) ( map (* (fromIntegral $ length x)) (average x)) (sumVect x) ) Thanks again Ondrej > On 2015-02-15 at 10:52, Ondrej Nekola <ondra(a)nekola.cz> wrote: >> How can I generate list of lists, where all the included lists are of >> the same length? > The general approach is to make a newtype, the Arbitrary instance of > which does what you want. For example, QucickCheck includes newtypes > Positive[1] and NonNegative[2], that generate restricted numbers. > > For your case, I'd write something like (untested): > > ``` > newtype RectList a = RectList [[a]] > > instance Arbitrary a => Arbitrary RectList a where > arbitrary = do > -- length of inner lists > m <- getPositive <$> arbitrary > -- length of outer list, can be zero > n <- getNonNegative <$> arbitrary > RectList <$> replicateM n (vector m) > ``` > > vector here is another function provided by QuickCheck: [3]. > > Hope this helps. > > bergey > >> Link for some tutorial *) can be, I believe, enough but I would be glad >> for any help. >> >> Thanks >> Ondra @satai Nekola >> ondra(a)nekola.cz >> >> *) Maybe I have not trained my google good enough yet, but lack of >> tutorials describing more than the most basic usecase is probably the >> most problematic point in re-learning Haskell so far. >> _______________________________________________ >> Beginners mailing list >> Beginners(a)haskell.org >> http://mail.haskell.org/cgi-bin/mailman/listinfo/beginners > Footnotes: > [1] http://hackage.haskell.org/package/QuickCheck-2.7.6/docs/Test-QuickCheck.ht… > > [2] http://hackage.haskell.org/package/QuickCheck-2.7.6/docs/Test-QuickCheck.ht… > > [3] http://hackage.haskell.org/package/QuickCheck-2.7.6/docs/Test-QuickCheck-Ar… > >

1 0

Spec with QuickCheck generationg list of list of the same length
by Ondrej Nekola 15 Feb '15

15 Feb '15

I am playing with Test.Hspec and Test.QuickCheck and I am traying to implement KMeans algorithm as a training. I use very simple representation of vectors: type LVector=[Double] I have no problems to write testcases with some concreate examples of input and expected output: it "sum of vectors is computed field by field" $ do (*sumVect* [[1.0, 2.0], [3.0, 4.0], [4, 5]]) `shouldBe` [8.0, 11.0] or simple properties based tests: it "zeroVector contains just zeroes" $ property (\(Positive d) -> [] == filter (/= 0.0) (*zeroVector*d) ) but I get sticked, when I try to create QuickCheck test, for example for the first case (*sumVect*). How can I generate list of lists, where all the included lists are of the same length? Link for some tutorial *) can be, I believe, enough but I would be glad for any help. Thanks Ondra @satai Nekola ondra(a)nekola.cz *) Maybe I have not trained my google good enough yet, but lack of tutorials describing more than the most basic usecase is probably the most problematic point in re-learning Haskell so far.

1 0

Getting "Discrete Mathematics Using a Computer" code to run
by Lawrence Bottorff 15 Feb '15

15 Feb '15

Saw this message: Error Loading Stdm.lhs in Haskell platform 2012 *Iwan Awaludin* awaludin at gmail.com <beginners%40haskell.org?Subject=Re%3A%20%5BHaskell-beginners%5D%20Error%20Loading%20Stdm.lhs%20in%20Haskell%20platform%202012&In-Reply-To=%3CCAHd5so7N7nX%3Ddiyk9Sv3%2B553gvqmXhWqGfgcHmAh%3D5XR0nHp8w%40mail.gmail.com%3E> *Tue Sep 4 07:08:20 CEST 2012* and that helped past that error. But now I get [1 of 1] Compiling Stdm ( .../Discrete/Stdm02142015.lhs, interpreted ) .../Discrete/Stdm02142015.lhs:1036:40-46: Not in scope: `catch' .../Discrete/Stdm02142015.lhs:1048:40-46: Not in scope: `catch' Failed, modules loaded: none. This is on ghci 7.6.3 lb

2 1

tower of hanoi problem
by Roelof Wobben 14 Feb '15

14 Feb '15

5 7

Can't connect to or contact haskell.org
by Federico Leva (Nemo) 12 Feb '15

12 Feb '15

Multiple bug reports inside. == 0. Original problem == Open https://www.haskell.org/cabal/users-guide/ == 1. Networking issue == The website doesn't load, I can't connect to www.haskell.org. Some other websites work, for instance --- community.haskell.org ping statistics --- 4 packets transmitted, 4 received, 0% packet loss, time 3004ms rtt min/avg/max/mdev = 27.645/31.075/35.602/2.886 ms You can see my IP in the headers of this email; I'm a Fastweb user in Italy and the traceroute is the following: $ traceroute haskell.org traceroute to haskell.org (23.253.242.70), 30 hops max, 60 byte packets 1 192.168.1.254 (192.168.1.254) 1.172 ms 6.645 ms 6.500 ms 2 10.57.94.195 (10.57.94.195) 9.440 ms 9.391 ms 9.371 ms 3 * * 10.251.136.89 (10.251.136.89) 16.638 ms 4 10.251.132.44 (10.251.132.44) 16.565 ms 16.550 ms 16.541 ms 5 * * * 6 10.251.137.186 (10.251.137.186) 16.661 ms 3.071 ms 11.863 ms 7 10.0.56.69 (10.0.56.69) 11.822 ms 11.857 ms 11.877 ms 8 10.0.1.94 (10.0.1.94) 17.569 ms 17.996 ms * 9 10.254.2.174 (10.254.2.174) 20.047 ms 20.015 ms 22.521 ms 10 10.254.9.210 (10.254.9.210) 22.463 ms 13.660 ms 13.531 ms 11 10.254.9.82 (10.254.9.82) 16.350 ms 16.370 ms 15.166 ms 12 10.1.128.38 (10.1.128.38) 16.620 ms 17.120 ms 17.009 ms 13 10.1.146.186 (10.1.146.186) 15.362 ms 15.352 ms 15.707 ms 14 * * * 15 * * * 16 * * * 17 * * * 18 * * * 19 * * * 20 * * * 21 * * * 22 * * * 23 * * * 24 * * * 25 * * * 26 * * * 27 * * * 28 * * * 29 * * * 30 * * * == 2. Contacts issue == I looked for contact information on your website (well, http://web.archive.org/web/20150102193702/https://www.haskell.org/haskellwi… ), but couldn't find any. http://code.haskell.org/~dons/haskell-1990-2000/msg06755.html still applies == 3. Trac registration form == http://trac.haskell.org is accessible for me, so I tried that instead. When I register, I get the following: Notice: An email has been sent to federicoleva@tiscali with a token to verify your new email address Oops… Trac detected an internal error: OperationalError: disk I/O error There was an internal error in Trac. It is recommended that you inform your local Trac administrator and give him all the information he needs to reproduce the issue. To that end, you could a ticket at this site. The action that triggered the error was: POST: /login == 4. Trac confirmation email == The confirmation email was never received, so I can't complete the registration the obvious "Error: Forbidden TICKET_CREATE privileges are required to perform this operation" ensues. I tried another address (this) as well, but that wasn't enough. == 5. Mailman? == So far the email interface to mailman worked, let's hope this gets through. :) If this isn't the appropriate place for the reports, please forgive my insufficient browse-by-proxy capabilities. Nemo

2 1

svgcairo/ghc vis cabal install error..
by madhub 11 Feb '15

11 Feb '15

Hi !, I tried installing svgcairo ( for installing ghc vis ). I am getting following error. Could you please help me out install ghc vis. Thanks in advance Madhu — environment details. I am on mac yosomite with gcc provided by OS X. Following is the gcc —version 22:54:34:~$ gcc --version Configured with: --prefix=/Applications/Xcode.app/Contents/Developer/usr --with-gxx-include-dir=/usr/include/c++/4.2.1 Apple LLVM version 6.0 (clang-600.0.56) (based on LLVM 3.5svn) Target: x86_64-apple-darwin14.1.0 Thread model: posix export PKG_CONFIG_PATH=/usr/local/lib/pkgconfig:/opt/X11/lib/pkgconfig export LDFLAGS=-L/usr/local/lib export CFLAGS=-I/usr/local/include export LIBS=-L/usr/local/lib svgcairo error =============== Configuring svgcairo-0.13.0.1... Building svgcairo-0.13.0.1... Preprocessing library svgcairo-0.13.0.1... gtk2hsC2hs: Error in C header file. /usr/include/sys/qos.h:124: (column 34) [FATAL] >>> Syntax error! The symbol `__attribute__' does not fit here. cabal: Error: some packages failed to install: svgcairo-0.13.0.1 failed during the building phase. The exception was: ExitFailure 1

2 1

Question re "Pattern match(es) are non-exhaustive"
by Joel Neely 11 Feb '15

11 Feb '15

I understand the meaning of the warning, but got it in a situation I didn't expect. The following source file contains the simplest case I could construct quickly to illustrate my question: Why does the first function (sumDigits) get the "non-exhaustive" warning? It contains a definition for both empty and non-empty arguments, just as the second (sumList), which does not get a warning. {-# OPTIONS_GHC -Wall #-} sumDigits :: [Integer] -> Integer sumDigits [] = 0 sumDigits (n:ns) | n < 10 = n + sumDigits ns | n >= 10 = r + sumDigits (q : ns) where (q, r) = n `quotRem` 10 sumList :: [Integer] -> Integer sumList [] = 0 sumList (n:ns) = n + sumList ns For completeness, here is the ghci transcript, with the location reported: Prelude> :load sumDigits.hs [1 of 1] Compiling Main ( sumDigits.hs, interpreted ) sumDigits.hs:4:1: Warning: Pattern match(es) are non-exhaustive In an equation for ‘sumDigits’: Patterns not matched: _ : _ Ok, modules loaded: Main. Thanks in advance for any guidance on this. -jn- -- Beauty of style and harmony and grace and good rhythm depend on simplicity. - Plato

5 10

Fwd: GHC 7.10 Prelude: we need your opinion
by Kim-Ee Yeoh 11 Feb '15

11 Feb '15

Simon PJ missed out the haskell-beginners list, but I encourage you to fill out the survey below as well. Now you may ask, "Well, I'm still learning haskell and I don't know enough." But consider that by being on this list and by haskelling whatever you may consider as utterly insignificant, you're still waaaaaaaay ahead than those who have yet to reach haskell at all. Your opinion counts. Because those _after_ you will help provide jobs by creating haskell projects, just like you. But unlike them, you get to influence the haskell of tomorrow that will be their first encounter, whether that would be a haskell that's friendlier or more @#$%^& than what it is today. The survey link is here: *http://goo.gl/forms/XP1W2JdfpX <http://goo.gl/forms/XP1W2JdfpX>* Summaries (each quite long) are here: · Overall summary: https://ghc.haskell.org/trac/ghc/wiki/Prelude710 <https://www.google.com/url?q=https%3A%2F%2Fghc.haskell.org%2Ftrac%2Fghc%2Fw…> · Details of Plan List: https://ghc.haskell.org/trac/ghc/wiki/Prelude710/List <https://www.google.com/url?q=https%3A%2F%2Fghc.haskell.org%2Ftrac%2Fghc%2Fw…> · Details of Plan FTP: https://ghc.haskell.org/trac/ghc/wiki/Prelude710/FTP <https://www.google.com/url?q=https%3A%2F%2Fghc.haskell.org%2Ftrac%2Fghc%2Fw…> Fwiw, I voted to delay FTP. In fact, here's what I wrote: "FTP is rushed and potentially unstable. I don't think we understand all about Traversable that we think we do, which might mean another big Prelude overhaul in the future. Actually, FTP doesn't even belong in 7.12 per se. It belongs in a new Haskell language standard and if that coincides with GHC 7.12, great. If not, it can wait for GHC 8.x and up." Because I've been keeping tabs on FTP I wrote my opinion without perusing the summaries, but on doing so later, I stumbled upon the following points that happen to reflect my exact observations, points that are relevant to learning haskell and growing the haskell community: https://ghc.haskell.org/trac/ghc/wiki/Prelude710/List - The existing corpus of books, tutorials, syllabi, and the like usually have a significant portion of the text dedicated to these very Prelude functions - and they would all need significant revision. - Teaching beginners what sequence means in its full generality is going to be a challenge. - While teaching beginners who end up on #haskell IRC might be possible, this is likely to increase the "bounce" rate, people who see Haskell, play around, and run away scared. I think Haskell probably has a higher bounce rate than most other languages, making it worse would be bad. Whether you agree or disagree with the above, now's the time to make yourself heard. The survey closes Feb 21. p.s. Needless to say, feel free to discuss FTP on this list should your opinion need a sounding board and a bit of time (not too long!) to firm up. -- Kim-Ee ---------- Forwarded message ---------- From: Simon Peyton Jones <simonpj(a)microsoft.com> Date: Tue, Feb 10, 2015 at 10:50 PM Subject: GHC 7.10 Prelude: we need your opinion To: "haskell(a)haskell.org" <haskell(a)haskell.org>, "Haskell Cafe ( haskell-cafe(a)haskell.org)" <haskell-cafe(a)haskell.org>, GHC users < glasgow-haskell-users(a)haskell.org>, "ghc-devs(a)haskell.org" < ghc-devs(a)haskell.org> Haskell Friends *This email asks for your help in deciding how to proceed with some Prelude changes in GHC 7.10. Please read on, but all the info is also at the survey link, here: http://goo.gl/forms/XP1W2JdfpX <http://goo.gl/forms/XP1W2JdfpX>. Deadline is 21 Feb.* The Core Libraries Committee (CLC) is responsible for developing the core libraries that ship with GHC. This is an important but painstaking task, and we owe the CLC a big vote of thanks for taking it on. For over a year the CLC has been working on integrating the *Foldable and Traversable classes* (shipped in base in GHC 7.8) into the core libraries, and into the Prelude in particular. Detailed planning for GHC 7.10 started in the autumn of 2014, and the CLC went ahead with this integration. Then we had a failure of communication. As these changes affect the Prelude, which is in scope for all users of Haskell, these changes should be held to a higher bar than the regular libraries@ review process. However, the Foldable/Traversable changes were not particularly well signposted. Many people have only recently woken up to them, and some have objected (both in principle and detail). This is an extremely unfortunate situation. On the one hand we are at RC2 for GHC 7.10, so library authors have invested effort in updating their libraries to the new Prelude. On the other, altering the Prelude is in effect altering the language, something we take pretty seriously. We should have had this debate back in 2014, but here we are, and it is unproductive to argue about whose fault it is. We all share responsibility. We need to decide what to do now. A small group of us met by Skype and we've decided to do this: · Push back GHC 7.10's release by at least a month, to late March. This delay also gives us breathing space to address an unrelated show-stopping bug, Trac #9858. · Invite input from the Haskell community on which of two approaches to adopt (this survey <http://goo.gl/forms/XP1W2JdfpX>). The main questions revolve around impact on the Haskell ecosystem (commercial applications, teaching, libraries, etc etc), so we want to ask your opinion rather than guess it. · Ask Simon Marlow and Simon Peyton Jones to decide which approach to follow for GHC 7.10. Wiki pages have been created summarizing these two primary alternatives, including many more points and counter-points and technical details: · Overall summary: https://ghc.haskell.org/trac/ghc/wiki/Prelude710 <https://www.google.com/url?q=https%3A%2F%2Fghc.haskell.org%2Ftrac%2Fghc%2Fw…> · Details of Plan List: https://ghc.haskell.org/trac/ghc/wiki/Prelude710/List <https://www.google.com/url?q=https%3A%2F%2Fghc.haskell.org%2Ftrac%2Fghc%2Fw…> · Details of Plan FTP: https://ghc.haskell.org/trac/ghc/wiki/Prelude710/FTP <https://www.google.com/url?q=https%3A%2F%2Fghc.haskell.org%2Ftrac%2Fghc%2Fw…> This survey invites your input on which plan we should follow. Would you please · Read the details of the alternative plans on the three wiki pages above · Add your response to the survey <http://goo.gl/forms/XP1W2JdfpX> Please do read the background. Well-informed responses will help. Thank you! *DEADLINE: 21 February 2015* Simon PJ _______________________________________________ ghc-devs mailing list ghc-devs(a)haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs

5 4