June 2015 - Haskell-Cafe

Space leak with replicateM
by Ken Takusagawa II 16 Jun '15

16 Jun '15

In the following program, the function "test1" results in huge memory usage, but substituting "test2", which does essentially the same thing, does not. GHC 7.10.1, AMD64. Is there a different implementation of replicateM that avoids the space leak? module Main where { import Control.Monad; numbers :: [Int]; numbers=[1..200]; -- has a space leak test1 :: [[Int]]; test1 = replicateM 4 numbers; -- no space leak test2 :: [[Int]]; test2 = do { x1 <- numbers; x2 <- numbers; x3 <- numbers; x4 <- numbers; return [x1,x2,x3,x4]; }; main :: IO(); main = print $ length $ test1; } Thanks, --ken

3 2

Question about constraining functions to particular ADT constructors
by Román González 16 Jun '15

16 Jun '15

Hello there, I've been thinking on different approaches to constraint particular functions to a particular constructor of an ADT in order to reduce representation of invalid states. Say for example I've this types: data Address = Address { ... } newtype Email = Email String data Package = Package { ... } data EmailMsg = EmailMsg { ... } data User = RealUser Address | VirtualUser Email And I would like to implement two functions: deliverPackage :: User -> Package -> IO Bool sendEmail :: User -> EmailMsg -> IO () I would like to constraint both deliverPackage and sendEmail to receive only the semantically correct constructor of User. I know of an approach I could use, that is wrapping each constructor in it's own newtype, and create some smart constructor that way, that approach works, but I find it rather verbose. Is there any other well known approach to deal with this scenarios?

3 2

Link to BayHac presentation and code files.
by David Banas 16 Jun '15

16 Jun '15

Hi all, Great to have seen you all again at BayHac! I couldn’t find any reference to a central repository for presentation/code upload. So, here’s my stuff (“Cracking the Vigenere Cipher in Haskell”): https://github.com/capn-freako/BayHac_2015 Many thanks to Graham Hutton for permission to post his code, along with my own. Cheers, -db

1 0

returning a polymorphic function
by Paolino 15 Jun '15

15 Jun '15

Hello list, I'm trying to accomplish something along this line {-# LANGUAGE ScopedTypeVariables #-} f :: IO (a -> a) f = return id main = do (g :: a -> a) <- f print $ g 1 print $ g "ciao" Is it possible or I have to call f more than once to get different g's Thanks paolino

3 5

PhD student or postdoc position announcement
by Thiemann, Rene 15 Jun '15

15 Jun '15

2-year postdoc or 3-year PhD-student position at the University of Innsbruck ============================================================================ The Computational Logic research group at the University of Innsbruck has one open position funded by the FWF (Austrian science fund) via the START project „Certifying Termination and Complexity Proofs of Programs“. The project aims at increasing the reliability in current complexity and termination provers by independently checking the generated proofs. To this end, several analysis techniques will be formalized in the theorem prover Isabelle/HOL, with a focus on term rewriting and Haskell. For this project, we are looking for an enthusiastic young researcher with a background in computational logic. Knowledge of automated termination analysis, complexity analysis, or theorem proving would be an asset. Candidates with a strong theoretical background in related areas are also encouraged to apply. The PhD-student candidate must have a Master's or equivalent degree. Knowledge of German is not essential. The salary is determined by the FWF-funding scheme (3.546,00 EUR monthly gross salary for postdocs, and 2.024,90 EUR for PhD-students, cf. http://www.fwf.ac.at/en/research-funding/personnel-costs/) Applications (including a CV, a publication list (only for postdocs), and a letter of recommendation) may be emailed to the project leader René Thiemann (rene.thiemann at uibk.ac.at) no later than July 6, 2015. We plan to make decisions on these position until July 10. The preferred starting date is summer or autumn 2015. Informal inquiries are also welcome via email. The city of Innsbruck is superbly located in the beautiful surroundings of the Tyrolean Alps. The combination of the Alpine environment and urban life in this historic town provides a high quality of living. Further information is available from the following links: START Project: http://cl-informatik.uibk.ac.at/research/projects/certifying-termination-an… Institute of Computer Science: http://informatik.uibk.ac.at/ University of Innsbruck: http://www.uibk.ac.at/ City of Innsbruck: http://www.innsbruck.at/

1 0

Data.Hashable instance generation?
by Matti Nykänen 15 Jun '15

15 Jun '15

Hello, A practical question about getting hashing to work. Having used Data.Map before, I wanted to try Data.HashMap.Lazy instead: ---- {-# LANGUAGE DeriveGeneric #-} import Data.HashMap.Lazy as HM import GHC.Generics (Generic) import Data.Hashable data Colour = Red | Green | Blue deriving Generic instance Hashable Colour foo = HM.insert Red ---- The data and its instance definition are directly from the web page <https://hackage.haskell.org/package/hashable-1.2.1.0/docs/Data-Hashable.html>. However, I det the following error: ---- GHCi, version 7.6.3: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. [1 of 1] Compiling Main ( htest.hs, interpreted ) htest.hs:13:8: No instance for (hashable-1.1.2.5:Data.Hashable.Hashable Colour) arising from a use of `insert' Possible fix: add an instance declaration for (hashable-1.1.2.5:Data.Hashable.Hashable Colour) In the expression: insert Red In an equation for `foo': foo = insert Red Failed, modules loaded: none. ---- I do not know what is wrong or how I could fix it. Note that my Haskell had originally the older hashable-1.0.0 package, but I cabal-installed this hashable-1.1.2.5 because surfing the net suggested that it would be much better. Could it be the case that the instance goes somehow in the wrong class? -- Matti Nykänen

2 2

Haskell dev role in Strats at Standard Chartered London
by Don Stewart 15 Jun '15

15 Jun '15

I'm hiring another Haskell dev to join the Strats team in London. Details here: https://donsbot.wordpress.com/2015/06/15/haskell-dev-role-in-strats-at-stan…

1 0

Haskell in Vietnam?
by Tom Schrijvers 15 Jun '15

15 Jun '15

Dear all, I would like to get in touch with people learning, teaching or using Haskell in Vietnam. Can anyone on this list help me? Thanks, Tom -- prof. dr. ir. Tom Schrijvers Research Professor KU Leuven Department of Computer Science Celestijnenlaan 200A 3001 Leuven Belgium Phone: +32 16 327 830 http://people.cs.kuleuven.be/~tom.schrijvers/

2 1

Recognizing projects schemes
by Ruben Astudillo 14 Jun '15

14 Jun '15

I recently came across the following blog-post: http://blog.cleancoder.com/uncle-bob/2015/04/15/DoesOrganizationMatter.html It speaks a bit of simplicity, efficiency and stuff that isn't important. What is important at least to me was the concept of project scheme, summarized on the following phrase: ``And so this gets to the crux of the question that you were really asking. You were asking whether the time required to learn the organization scheme of the system is worth the bother. Learning that organization scheme is hard. Becoming proficient at reading and changing the code within that scheme take time, effort, and practice. And that can feel like a waste when you compare it to how simple life was when you only had 100 lines of code.'' And that is something I totally struggle at approaching new projects. The only reason I could understand XMonad for example is because they gave a general overview (thanks) of it on the Developing module. I feel I got a problem of methodology. What approaches are you guys using to understanding new projects schemes on a efficient manner? How long it usually takes you?. Any advices? Thanks in advance. -- Ruben Astudillo. pgp: 0x3C332311 , usala en lo posible :-) Crear un haiku, en diecisiete silabas, es complica...

2 1

ANN: TestExplode 0.1.0.0: testscript generator for all target languages
by Hans-Juergen Guth 13 Jun '15

13 Jun '15

My new Haskell program/library is released! What it does: ------------- Instead of writing one test-script after another, you define the structure of test-sets. Advantages: ----------- More test-scripts with less code, better overview, easier review of the testcases, easier change and addition of testcases. Details: -------- TestExplode generates testcases out of program snippets. The snippets are treated as text, with variables, that are subsituted (thanks the package perl6-interpolated-strings). The text-snippets are annotated and can form a directed acyclic graph. Each path and each variable substitution is a testcase. Subgraphs can be defined and used in many testcases. The graph can be printed with graphviz. The expected results of the testcase can partially defined by changing the state of the testcase by each testsnippet. So you can define a state-machine in haskell and the annotations to the text-snippets give the input to the state-machine, dependant of the test-input and which path through the graph is chosen. The program/library is designed with users in mind, who have little or no knowledge of haskell at the beginning. It's my first bigger haskell program, so it is probably relativ simple haskell code in the library and in the user interface. Examples can be found in the documentation. Further info: ------------- A "Getting Started": https://hackage.haskell.org/package/TestExplode-0.1.0.0/readme The Library which generates the testcases https://hackage.haskell.org/package/TestExplode-0.1.0.0/docs/TestExplode-Te… The Library which makes FGL-Graphs for the module grahpviz out of my own definition of directed acyclic graphs https://hackage.haskell.org/package/TestExplode-0.1.0.0/docs/TestExplode-Di… I would be happy about comments!

1 0