- Haskell-Cafe - Haskell.org

Float instance of 'read'
by Mauricio 17 Sep '08

17 Sep '08

Hi, A small annoyance some users outside english speaking countries usually experiment when learning programming languages is that real numbers use a '.' instead of ','. Of course, that is not such a problem except for the inconsistence between computer and free hand notation. Do you think 'read' (actually, 'readsPrec'?) could be made to also read the international convention (ie., read "1,5" would also work besides read "1.5")? I'm happy to finaly use a language where I can use words of my language to name variables, so I wonder if we could also make that step. Thanks, Maurício

13 20

Predicativity?
by Wei Hu 17 Sep '08

17 Sep '08

Hello, I only have a vague understanding of predicativity/impredicativity, but cannot map this concept to practice. We know the type of id is forall a. a -> a. I thought id could not be applied to itself under predicative polymorphism. But Haksell and OCaml both type check (id id) with no problem. Is my understanding wrong? Can you show an example that doesn't type check under predicative polymorphism, but would type check under impredicative polymorphism? Thanks!

3 4

Semantic Domain, Function, and denotational model.....
by Greg Meredith 17 Sep '08

17 Sep '08

Daryoush, Hopefully, the other replies about proving the monad laws already answered your previous question: yes! As for notions of semantic domain and denotational model, these ideas go back quite a ways; but, were given solid footing by Dana Scott<http://en.wikipedia.org/wiki/Dana_Scott>. In a nutshell, we have essentially three views of a computation - Operational <http://en.wikipedia.org/wiki/Operational_semantics> -- computation is captured in a program and rules for executing it - Logical <http://en.wikipedia.org/wiki/Proof-theoretic_semantics> -- computation is captured in a proof and rules for normalizing it - Denotational <http://en.wikipedia.org/wiki/Denotational_semantics> -- computation is captured as a (completely unfolded) mathematical structure In the latter view we think of computations/programs as denoting some (usually infinitary) mathematical object. Our aim<http://en.wikipedia.org/wiki/Domain_theory>is to completely define the meaning of programs in terms of maps between their syntactic representation and the mathematical objects their syntax is intended to denote. (Notationally, one often writes such maps as [| - |] : Program -> Denotation.) For example, we model terms in the lambda calculus as elements in a D-infinity domain or Bohm trees or ... . Or, in more modern parlance, we model terms in the lambda calculus as morphisms in some Cartesian closed category, and the types of those terms as objects in said category. The gold standard of quality of such a model is full abstraction<http://en.wikipedia.org/wiki/Denotational_semantics#Full_abstraction>-- when the denotational notion of equivalence exactly coincides with an intended operational notion of equivalence. In symbols, - P ~ Q <=> [| P |] = [| Q |], where ~ and = are the operational and denotational notions of equivalence, respectively The techniques of denotational semantics have been very fruitful, but greatly improved by having to rub elbows with operational characterizations. The original proposals for denotational models of the lambda calculus were much too arms length from the intensional structure implicit in the notion of computation and thus failed to achieve full abstraction even for toy models of lambda enriched with naturals and booleans (cf the so-called full abstraction for PCF problem<http://en.wikipedia.org/wiki/Programming_language_for_Computable_Functions>). On the flip side, providing a better syntactic exposure of the use of resources -- ala linear logic -- aided the denotational programme. More generally, operational models fit neatly into two ready-made notions of equivalence - contextual equivalence<http://encyclopedia2.thefreedictionary.com/Contextual+equivalence>-- behaves the same in all contexts - bisimulation <http://en.wikipedia.org/wiki/Bisimulation> -- behaves the same under all observations Moreover, these can be seen to coincide with ready-made notions of equivalence under the logical view of programs. Except for syntactically derived initial/final denotational models -- the current theory usually finds a more "home-grown" denotational notion of equivalence. In fact, i submit that it is this very distance from the syntactic presentation that has weakened the more popular understanding of "denotational" models to just this -- whenever you have some compositionally defined map from one representation to another that respects some form of equivalence, the target is viewed as the denotation. Best wishes, --greg Date: Mon, 15 Sep 2008 10:13:53 -0700 From: "Daryoush Mehrtash" <dmehrtash(a)gmail.com> Subject: Re: [Haskell-cafe] Semantic Domain, Function, and denotational model..... To: "Ryan Ingram" <ryani.spam(a)gmail.com> Cc: Haskell Cafe <haskell-cafe(a)haskell.org> Message-ID: <e5b8e9790809151013k53fdc105i88f2f47f1f9b16bd(a)mail.gmail.com> Content-Type: text/plain; charset=ISO-8859-1 Interestingly, I was trying to read his paper when I realized that I needed to figure out the meaning of denotational model, semantic domain, semantic functions. Other Haskell books didn't talk about design in those terms, but obviously for him this is how he is driving his design. I am looking for a simpler tutorial, text book like reference on the topic. Daryoush On Mon, Sep 15, 2008 at 1:33 AM, Ryan Ingram <ryani.spam(a)gmail.com> wrote: > I recommend reading Conal Elliott's "Efficient Functional Reactivity" > paper for an in-depth real-world example. > > http://www.conal.net/papers/simply-reactive > > -- ryan > > On Sun, Sep 14, 2008 at 11:31 AM, Daryoush Mehrtash <dmehrtash(a)gmail.com> wrote: >> I have been told that for a Haskell/Functional programmer the process >> of design starts with defining Semantic Domain, Function, and >> denotational model of the problem. I have done some googling on the >> topic but haven't found a good reference on it. I would appreciate >> any good references on the topic. >> >> thanks, >> >> daryoush >> >> ps. I have found referneces like >> http://en.wikibooks.org/wiki/Haskell/Denotational_semantics which >> talks about semantic domain for "the Haskell programs 10, 9+1, 2*5" >> which doesn't do any good for me. I need something with a more real >> examples. >> _______________________________________________ >> Haskell-Cafe mailing list >> Haskell-Cafe(a)haskell.org >> http://www.haskell.org/mailman/listinfo/haskell-cafe >> > -- L.G. Meredith Managing Partner Biosimilarity LLC 806 55th St NE Seattle, WA 98105 +1 206.650.3740 http://biosimilarity.blogspot.com

2 2

Mac OS X dylib woes
by John MacFarlane 17 Sep '08

17 Sep '08

I'm hoping some Haskell developers who use Macs can help me with this one. I can install pcre-light just fine using cabal install. But when I try to use it, I get this error: GHCi, version 6.8.2: http://www.haskell.org/ghc/ :? for help :Loading package base ... linking ... done. Prelude> :m Text.Regex.PCRE.Light.Char8 Prelude Text.Regex.PCRE.Light.Char8> compile "h." [] Loading package array-0.1.0.0 ... linking ... done. Loading package bytestring-0.9.0.1 ... linking ... done. Loading package pcre-light-0.3.1 ... can't load .so/.DLL for: pcre (dlopen(libpcre.dylib, 10): image not found) OK, so it can't find the pcre library (which is in /opt/local/lib). I can fix that: export DYLD_LIBRARY_PATH=/opt/local/lib Now it works. But other things are broken! For example, I can't run vim, which looks for a library called libJPEG.dylib and now finds libjpeg.dylib in /opt/local/lib (case-insensitive file system!). The apple website recommends against setting DYLD_LIBRARY_PATH. Instead, they say, the paths to the libraries should be hard-coded into the binary: http://discussions.apple.com/thread.jspa?threadID=1670523&tstart=0 I'm sure others have had the same problem. What's the solution? John

3 3

FRP (was: Semantic Domain, Function, and denotational model)
by John Lato 17 Sep '08

17 Sep '08

I just noticed that the "Simply Efficient Functional Reactivity" paper has been updated since I last looked; I'll have to read it again now. Is the library/code mentioned in the paper released or available anywhere at this time? Conal has left tantalizing hints scattered in various places... Thanks, John Lato

3 2

GHC trouble on Leopard
by Miguel Vilaca 17 Sep '08

17 Sep '08

Hi, I tried to compile some code on Mac Os X (Intel) Leopard. I have GHC 6.8.3 installed - the installer from GHC webpage (GHC-6.8.3- i386.pkg). But when I run make I get this error ghc-6.8.3: could not execute: /Library/Frameworks/GHC.framework/ Versions/608/usr/lib/ghc-6.8.3/ghc-asm The work-around of removing the option -fvia-C pointed somewhere on the web is not an option here due to library dependecies. Any hints on how to solve this? Thanks in advance Miguel Vilaça

3 3

Hyena Status? Re: [Haskell-cafe] Starting Haskell with a web application
by Donnie Jones 17 Sep '08

17 Sep '08

Hello Johan Tibell, Hyena looks very interesting. From the github tracking, you've been working... Maybe a release soon? Also, I saw your slides from the 'Left-fold enumerators' presentation at Galois. Maybe include the slides in the docs/ for a release? Thank you. __ Donnie On Thu, Mar 6, 2008 at 7:40 AM, Johan Tibell <johan.tibell(a)gmail.com> wrote: > > Do you (both) have repos that I could download from? I quite interested > > in both projects, esp. the WSGI clone. > > Yes and no. The code [1] is in my darcs repository but is in an > unusable state until I've fixed my incremental parser (in > Hyena/Parser.hs) which I plan to do next week. I would like the first > release to be nice and polished so I'm trying to not release anything > prematurely. > > 1. http://darcs.johantibell.com/hyena/ > _______________________________________________ > Haskell-Cafe mailing list > Haskell-Cafe(a)haskell.org > http://www.haskell.org/mailman/listinfo/haskell-cafe >

2 2

Problem with hscurses
by david48 17 Sep '08

17 Sep '08

the getCh funtion is supposed to return an interpreted Key with values like KeyChar c, KeyReturn, KeyBackspace, etc. but in fact, it only ever returns KeyChar c values ! am I doing anything wrong ? Here's an example program : module Main where import UI.HSCurses.Curses import Text.Printf import System.IO main = do hSetBuffering stdout NoBuffering withCurses test1 withCurses f = do initScr echo False cBreak True res <- f endWin return res test1 = do c <- getCh printf "Touche : %s\r\n" (show c) test1

1 1

hircules resurrected (partially ;-)
by wman 17 Sep '08

17 Sep '08

I stumbled upon hircules-0.3 IRC client and decided to take it for a test drive. I managed to get it to compile under ghc-6.8.3/win32, converted it to use Text.Codec.Iconv (the old version of included Iconv.hs should also work), but it seems it doesn't work. It only gives an empty form and according to wireshark doesn't even attempt a connection. If somebody knew hircules in it's days of glory and/or wants to go bug-hunting, i'll gladly provide the modified version, as I probably won't spend more time on it in near future (at least until i get to know gtk2hs better).

1 0

Comparing GADTs for Eq and Ord
by Tom Hawkins 17 Sep '08

17 Sep '08

Hi, How do I compare a GADT type if a particular constructor returns a concrete type parameter? For example: data Expr :: * -> * where Const :: Expr a Equal :: Expr a -> Expr a -> Expr Bool -- If this read Expr a, the compiler has no problem. instance Eq (Expr a) where Const == Const = True (Equal a b) (Equal x y) = a == x && b == y _ == _ = False GHC throws: Couldn't match expected type `a1' against inferred type `a2' `a1' is a rigid type variable bound by the constructor `Equal' at Test.hs:9:3 `a2' is a rigid type variable bound by the constructor `Equal' at Test.hs:9:18 Expected type: Expr a1 Inferred type: Expr a2 In the second argument of `(==)', namely `x' In the first argument of `(&&)', namely `a == x' I believe I understand the reason for the problem; even though Equal has a type Expr Bool, two Equal expressions could have parameters of different types. But I'm not sure how to get around the problem. Any suggestions? Thanks, Tom

7 12