- Haskell-Cafe - Haskell.org

Re: [Haskell-cafe] Why are field selectors functions?
by David Banas 09 Aug '13

09 Aug '13

Wouldn't the implementation hiding feature of the *newtype *idiom be broken, if field selectors were not first class functions? For instance, the following code (taken shamelessly from Ch. 10 of *Real World Haskell*): module Parse ( runParser ) where data ParseState = ParseState { string :: String } deriving (Show) newtype Parser a = Parser { runParser :: ParseState -> Either String (a, ParseState) } has the attractive feature of hiding the internal implementation of the *ParseState *and *Parser *types from the user, preventing him from, for instance, pattern matching on either and thus writing code, which may break when we change the implementation. I believe this is only possible, because the *runParser *accessor is exportable as a first class function.

1 0

ANN: hi2 -- a better indentation mode for Emacs' haskell-mode
by Gergely Risko 09 Aug '13

09 Aug '13

Hi, In the last 2-3 weeks I've been working on Haskell indentation inside Emacs. I had some annoyances for a long time and fixed some of them. The new mode is called hi2, it's heavily based on the current haskell-indentation (part of haskell-mode). The changes are mainly to the UI, although, I plan to have a look on the parser too. The code can be found on github: https://github.com/errge/hi2. Feel free to send me feedback, bug reports, pull requests, etc. It's also hosted on MELPA. The most notable changes so far are the following: - DEL and C-d is not mapped: if you want to indent backwards, you can use S-TAB. This means no random jumping on backspace anymore. - TAB steps to the right as before, but when the end is reached, it starts going to the left instead of wrapping around. - TAB stays inside the code, instead of going to the beginning of the line. As in pyhton-mode and perl-mode. - Region indentation common case is supported: TAB and S-TAB is simply moving the whole region to the left/right by 1-column. Can be pressed repeatedly. - The current indentations are shown as underscores in the current line. So you have some visual indication on what's gonna happen. Also useful while hacking on the parser and want to see the results. Can be turned off by setting hi2-show-indentations to nil in your init file or calling hi2-disable-show-indentations from the buffer. If there are collisions with other overlay hacking modes (e.g. fill-column-indicator), try to turn off hi2-show-indentations-after-eol. - The buffer is not changed when indentation is not changed (so there are no undo points created and no dirty flag in the buffer if pressing TAB had no effect). - The code for all this is somewhat commented and cleaned. I'm not trying to fork haskell-mode, it's maintained very actively on github and the maintainers were always friendly through code reviews. But the current haskell-indentation-mode required so many fixes that it became unpleasant to contribute patch-by-patch, I wanted to move fast. When (and if) hi2 gets stable and better in every respect than haskell-indentation-mode, I'm happy to merge and contribute it back. Gergely

2 1

starting GHC development -- two questions
by Ömer Sinan Ağacan 09 Aug '13

09 Aug '13

Hi all, I want to start working on GHC, I cloned the repo, made some trivial changes(change some strings etc.) and re-built and observed the results. Now I'll continue reading and understanding the source. While doing this, I think one feature would greatly help me finding my way through GHC source, which is huge: I want to see definition of some symbols. Normally what I would do for this is to load source into GHCi and run :info command. But in the case of GHC, even if it's possible to load GHC into GHCi, I don't think it will be faster than running "ack --haskell someSymbol" and searching through results manually. First idea came to my mind was to generate tags files and then navigate from within vim(my editor of choice). tags file can be added to Makefile as a goal and then tags can be regenerated after each build. Did anyone else try this before? My second question is do we have any low-hanging fruits in trac, to help new people start contributing to GHC? I know several open source projects using that approach and it's really helpful for beginners. I just skimmed over trac and most issues look way too advanced for a starter. Thanks in advance, --- Ömer Sinan Ağacan http://osa1.net

3 3

Errors with Template Haskell
by Jose A. Lopes 09 Aug '13

09 Aug '13

Hi, In Template Haskell, what is the proper way of signalling an error ? For example, you are generating code and you detect that a given parameter does not fulfill a precondition (e.g., String is empty), and you want to abort compilation with a descriptive error message. Thanks, Jose -- Jose Antonio Lopes Ganeti Engineering Google Germany GmbH Dienerstr. 12, 80331, München Registergericht und -nummer: Hamburg, HRB 86891 Sitz der Gesellschaft: Hamburg Geschäftsführer: Graham Law, Christine Elizabeth Flores Steuernummer: 48/725/00206 Umsatzsteueridentifikationsnummer: DE813741370

2 2

Re: [Haskell-cafe] Haskell Platform and Leksah on Windows
by Carter Schonwald 09 Aug '13

09 Aug '13

Hey Dorin, I don't understand your claims. 1) haskell has worked perfectly well on windows for quite some time. I used HUGs nearly a decade ago, and in more recent time (2-3 years ago) I helped teach an introductory first computer science class using GHC where many students were doing great work using notepad++ and ghci. I don't understand your focus on emacs and make files. 2) if you want an "IDE" experience, Sublime Text with the right plugins, or perhaps EclipseFP are worth checking out. 3) likewise, if you're finding tooling on windows unsatisfactory, help fix it! Bug reports, patches, or new tools and libraries are always welcome. Haskell is a relatively small community, and thusly limited manpower (we're all volunteers), so way to fix any problem is help out! cheers On Thu, Aug 8, 2013 at 3:30 AM, Dorin Lazar <dorin.lazar(a)gmail.com> wrote: > Hello, > I am the original author of the post, and I finally received the > emails from the mailman (probably there was an issue with the > automated requests). > My answers are inlined. > > > 1) Leksah should not be considered an "official haskell ide", but merely > one of > > many community supported editing tools. And frankly one of the less > widely > > used ones at that! Leksah is not used much at all by anyone, though > theres > > probably a handful of folks who do use it. > > Many folks use editors like Sublime Tex (2/3), Emacs, Vi(m), textmate, > and > > many more. Its worth noting that the sublime-haskell plugin for sublime > > text, and analogous packages for many other editors, provide haskell > > IDE-like powers, or at least a nice subset thereof. > Unfortunately, I think the problem with this is that we have a > different vision on how development should be done. I have extensive > experience of working from console, with a simple text editor and > hand-made Makefiles or anything similar. However, an IDE should be a > productivity tool, that can help you improve your understanding of the > language, and can assist you in following the proper syntax for a new > language. While learning by doing 'write, save, compile, examine error > message' is ok with me, it is slow, and it limits the time I can > dedicate to learning the language itself. A better cycle is the > current 'write, examine error message' of most IDEs, since it's faster > and requires no context switch. Sure, editors can help there. IDEs do > this by default. > So it's normal of me to search for an IDE to better learn the > language, I'll leave the emacs + console version for when I am > productive in the language. > > > 2) There are people working on building better easily portable native gui > > toolkits, but in many respects, a nice haskelly gui toolkit is still > > something people are experimetning with how to do well. theres lots of > great > > tools out as of the past year or two, many more in progress on various > time > > scales, and gtk2hs is great for linux (and thats fine). > Unfortunately, this is not what's advertised. In fact, on the leksah > site, the recommended method is to have the IDE installed via cabal. > In another mail Mihai calls me unreasonable, but I think it's > reasonable to think that the recommended method should be the one that > works. > But the easy to tell truth is that the Haskell Platform for Windows > is not mature enough yet. That is something I can understand, and I > can recommend other beginners to install a Linux VM for Haskell. That > is perfectly fine, zero cost, 100% gain. However, the mistakes from > the Haskell Platform as it is now on Windows should be pointed out, > and although I've been called a mystical animal that wants only free > support, I think what I had in that blog post was actually a bug > report for the people that can actually add 1+1 to make 2 when it > comes to the Haskell Platform for Windows. Surely, I was harsh. But > that's the first experience of a beginner with Haskell, and I chose to > contribute my experience to people more knowledgeable instead of > shutting up and hiding the dust under the rug. > > Many thanks, > Dorin >

5 4

deriving Data.HashTable - stack overflow
by Lyle Kopnicky 09 Aug '13

09 Aug '13

Hi folks, In GHC 7.6.3, the base Data.HashTable is deprecated, so I installed the hashtables package. In order to work on your datatype, you need an instance of Data.Hashable.Hashable. So I went to the Data.Hashable page and looked up examples on how to derive a Hashable instance for my datatype: http://hackage.haskell.org/packages/archive/hashable/latest/doc/html/Data-H… The problem occurs even when using the sample code on the page: {-# LANGUAGE DeriveGeneric #-} import GHC.Generics (Generic) import Data.Hashable data Colour = Red | Green | Blue deriving Generic instance Hashable Colour If I then type `hash Red` I get a stack overflow. I am using the Haskell Platform, so I have hashable-1.1.2.5, but I notice the docs are for hashable-1.2.0.10. If I install 1.2.0.10 though, other code in my project breaks - seems like one part doesn't recognize the instances from another part. So I'll stick with the platform version. I am running the 32-bit Windows GHC 7.6.3. Haskell Platform 2013.2.0.0. Any thoughts? Thanks, Lyle

4 6

Haskell Platform and Leksah on Windows
by Mihai Maruseac 09 Aug '13

09 Aug '13

Hello, A friend of mine tried to install Haskell Platform and Leksah on Windows and was troubled by the amount of problems he encountered as a beginner in this. I've told him to ask over IRC and mailing list but it seems he has some problems with registration. Anyway, he blogged about his problems at http://dorinlazar.ro/haskell-platform-windows-crippled/ and I'm sure that we can work on fixing some of them. -- MM "All we have to decide is what we do with the time that is given to us"

7 8

Contracts
by Julian Arni 08 Aug '13

08 Aug '13

I've come across interesting papers - and old, sometimes inaccessible, repositories - related contracts in Haskell (Dana Xu and SPJ's papers; haskellcontracts and the Programatica Project). And then some newer, apparently not quite mature projects (Dan Rosen's repositories, for instance). Is there anything in between - mature, but not abandoned? Thanks, Julian

3 4

ANN: FunGEn 0.4.2, an easy cross-platform OpenGL/GLUT game engine
by Simon Michael 08 Aug '13

08 Aug '13

I'm pleased to announce the hackage release of FunGEn 0.4! (Actually 0.4.2 as my 0.4 announcement did not reach the mail lists.) FunGEn (Functional Game Engine) is a BSD-licensed, cross-platform, OpenGL/GLUT-based, imperative game engine/framework. With very few dependencies and two example games, it's one of the easiest ways to get started with game development in Haskell. FunGEn was probably the first Haskell game framework, created by Andre Furtado in 2002 (!). Here's his original feature list: * Initialization, updating, removing, rendering and grouping routines for game objects; * Definition of a game background (or map), including texture-based maps and tile maps; * Reading and intepretation of the player's keyboard input; * Collision detection; * Time-based functions and pre-defined game actions; * Loading and displaying of 24-bit bitmap files; * Some debugging and game performance evaluation facilities; * Sound support (actually for windows platforms only... :-[ ) What's new in 0.4.x: * a new hakyll-based website, incorporating the old site content * new haddock documentation * tested with GHC 7.6 * fixed buggy input when holding down keys on windows * input handlers now receive mouse position and modifier state (inspired by Pradeep Kumar; see fungentest.hs for examples) * added q as quit key in examples Home: http://joyful.com/fungen Hackage: http://hackage.haskell.org/package/FunGEn Code: http://hub.darcs.net/simon/fungen Install from hackage: $ cabal update $ cabal install FunGEn Install source and run examples: $ darcs get http://hub.darcs.net/simon/fungen $ cd fungen $ cabal install $ (cd examples/pong; ghc pong; ./pong) $ (cd examples/worms; ghc worms; ./worms) Contribute patches: - log in to hub.darcs.net and fork http://hub.darcs.net/simon/fungen - push changes to your branch - give me a "pull request" on #haskell-game I have maintained FunGEn very sporadically. If you'd like to take it and run with it, or co-maintain, let's chat! I'm sm on the #haskell-game IRC channel. -Simon

1 0

Installing wxHaskel on Snow Leopard
by Eduardo Sato 08 Aug '13

08 Aug '13

Hello, guys. Has anybody tried to install wxhaskell on Snow Leopard? I followed these instructions: http://www.haskell.org/haskellwiki/WxHaskell/Mac , but got an error: src/haskell/Graphics/UI/WXCore/WxcClassesAL.hs:13085:1: Unacceptable argument type in foreign declaration: CInt When checking declaration: foreign import ccall safe "static wxLogWindow_Create" wxLogWindow_Create :: Ptr (TWindow a) -> CWString -> CBool -> CBool -> IO (Ptr (TLogWindow ())) src/haskell/Graphics/UI/WXCore/WxcClassesAL.hs:13085:1: Unacceptable argument type in foreign declaration: CInt When checking declaration: foreign import ccall safe "static wxLogWindow_Create" wxLogWindow_Create :: Ptr (TWindow a) -> CWString -> CBool -> CBool -> IO (Ptr (TLogWindow ())) Failed to install wxcore-0.90.0.3 cabal: Error: some packages failed to install: wx-0.90.0.1 depends on wxcore-0.90.0.3 which failed to install. wxcore-0.90.0.3 failed during the building phase. The exception was: ExitFailure 1 http://lpaste.net/91634 I'm using ghc 7.6.3. I've heard the above problems are related to GHC 7.6.3 being more pedantic about FFI declarations. Is there any way around it? Thanks -- Eduardo Sato

4 7