March 2009 - Haskell-Cafe

Re: [Haskell-cafe] Dynamically typing TH.Exp at runtime
by Martin Hofmann 12 Mar '09

12 Mar '09

This is what hint does, isn't it? However, this still leaves it to me to infer what type [|a|] in [| reverse (a:b:[]) = (b:a:[]) |] is. Thanks a lot anyway. Martin On Thu, 2009-03-12 at 12:59 -0200, Daniel Gorín wrote: > would using ghc-as-a-library to type (strings with) haskell > expressions at runtime help? > > On Mar 12, 2009, at 12:37 PM, Martin Hofmann wrote: > > > I am doing meta-programming at runtime. So my program gets a full > > Haskell declaration in expression quotation ([d|...|]) modifies it and > > returns the modified expression. Therefore, I need type information of > > this expression, and any subexpression, at _runtime_ ! For example: > > > > [d| reverse x1 = y1 |] > > > > - rewrites_to -> > > > > [d| reverse x2:xs = y2:ys |] > > > > - rewrites_to -> > > > > [d| reverse x:xs = reverse xs ++ [x] |] > > > > reverse :: [a] -> [a] implies > > x2,y2 :: a > > x1,y1,xs,ys :: [a] > > > > TH.reify is not applicable, because I need the information at runtime > > and I am in IO. I suppose Data.Dynamic does not work either, because > > [| > > xs|] :: ExpQ and not [a]. > > > > So it looks like I need my own type checker and inference and tag each > > subexpression with its type. If so, I can even omit TH and use my own > > data type for the abstract syntax tree. This annoys me a bit, because > > for me it seems that all I need is already there. > > > > Did anybody have similar problems, because I shouldn't be the only one > > doing dynamic typing in a static language? Is there a Haskell > > implementation of the paper "Typing Dynamic Typing" by Baars and > > Swierstra (http://people.cs.uu.nl/arthurb/dynamic.html ), so I can try > > out if this helps me? > > > > Any help, comments, and how-to-suggestions are highly welcome. > > > > Thanks a lot, > > > > Martin > > > > > > > > _______________________________________________ > > Haskell-Cafe mailing list > > Haskell-Cafe(a)haskell.org > > http://www.haskell.org/mailman/listinfo/haskell-cafe

1 0

Data.Binary patches?
by Denis Bueno 12 Mar '09

12 Mar '09

I've got a small patch for Data.Binary. Should I post it here, or is there some more appropriate forum? http://code.haskell.org/binary/ doesn't specify. Thanks, Denis

3 3

"The Curse of Our Times" in different languages
by Jamie 12 Mar '09

12 Mar '09

See Black-Scholes model/option formula being implemented in different languages including Haskell. http://www.espenhaug.com/black_scholes.html Jamie

1 0

ThreadScope: Request for features for the performance tuning of parallel and concurrent Haskell programs
by Satnam Singh 12 Mar '09

12 Mar '09

Donnie Jones, Simon Marlow and I have been working on infrastructure for logging run-time events and a graphical viewer program called ThreadScope. Hopefully these features will make it into the next release of GHC. We hope the event-log viewer ThreadScope will be useful for the performance tuning of parallel and concurrent Haskell programs. You can see a few screen shots at the program's website http://raintown.org/threadscope Before making the release I thought it would be an idea to ask people what other features people would find useful or performance tuning. So if you have any suggestions please do let us know! Cheers, Satnam Singh ________________________________ Satnam Singh Microsoft 7 JJ Thomson Avenue Cambridge CB3 0FB United Kingdom Email: satnams(a)microsoft.com<mailto:satnams@microsoft.com> UK tel: +44 1223 479905 Fax: +44 1223 479 999 UK mobile: +44 7979 648412 USA cell: 206 330 1580 USA tel: 206 219 9024 URL: http://research.microsoft.com/~satnams Live Messenger: satnam(a)raintown.org<mailto:satnam@raintown.org>

4 3

monadic logo
by Gregg Reynolds 12 Mar '09

12 Mar '09

At risk of becoming the most hated man in all Haskelldom, I'd like to suggest that the Haskell logo not use lambda symbols. Or at least not as the central element. Sorry, I know I'm late to the party, but the thing is there is nothing distinctive about lambda; it's common to all FPLs. Besides, Lisp/Scheme already have that franchise. What is distinctive about Haskell it's use of the monad. The Pythagorean monad symbol is wonderfully simple: http://en.wikipedia.org/wiki/Monad_(Greek_philosophy)<http://en.wikipedia.org/wiki/Monad_%28Greek_philosophy%29>. Something might also be done with the triad to reflect the fact that the monad in cat theory is actually a triple ( http://en.wikipedia.org/wiki/Triad_(Greek_philosophy)<http://en.wikipedia.org/wiki/Triad_%28Greek_philosophy%29> ). The "Cup or Monad" (http://www.sacred-texts.com/chr/herm/hermes4.htm) with a little bit of work, could be turned into an amusing Haskell manifesto. Lots of interesting imagery from the gnostic tradition ( http://www.sacred-texts.com/gno/th2/index.htm) although SICP seems already to have used something similar ( http://mitpress.mit.edu/images/products/books/0262011530-f30.jpg) Hermes Trismegistus = Thrice-great Hermes -> Haskell Trismegistus, Thrice-Glorious Haskell, etc. Might be too cute. There's also Leibniz' monadology - I can't think of any visual imagery to go with it, but he did end up with the "best of all possible worlds" hypothesis, which gives us a slogan sure to irritate: "Haskell - the best of all possible languages". Not to mention "Haskell Monadology" as a name for the official Haskell definition, etc. Hey, at least it isn't cute. -gregg

3 2

Dynamically typing TH.Exp at runtime
by Martin Hofmann 12 Mar '09

12 Mar '09

I am doing meta-programming at runtime. So my program gets a full Haskell declaration in expression quotation ([d|...|]) modifies it and returns the modified expression. Therefore, I need type information of this expression, and any subexpression, at _runtime_ ! For example: [d| reverse x1 = y1 |] - rewrites_to -> [d| reverse x2:xs = y2:ys |] - rewrites_to -> [d| reverse x:xs = reverse xs ++ [x] |] reverse :: [a] -> [a] implies x2,y2 :: a x1,y1,xs,ys :: [a] TH.reify is not applicable, because I need the information at runtime and I am in IO. I suppose Data.Dynamic does not work either, because [| xs|] :: ExpQ and not [a]. So it looks like I need my own type checker and inference and tag each subexpression with its type. If so, I can even omit TH and use my own data type for the abstract syntax tree. This annoys me a bit, because for me it seems that all I need is already there. Did anybody have similar problems, because I shouldn't be the only one doing dynamic typing in a static language? Is there a Haskell implementation of the paper "Typing Dynamic Typing" by Baars and Swierstra (http://people.cs.uu.nl/arthurb/dynamic.html ), so I can try out if this helps me? Any help, comments, and how-to-suggestions are highly welcome. Thanks a lot, Martin

1 0

Request: warn about language extensions that are not used
by Peter Verswyvelen 12 Mar '09

12 Mar '09

When I put {-# OPTIONS_GHC -Wall -Werror #-} in my source file, I don't get compiler (GHC) warnings about redundant language extensions that I enabled. It would be nice if the compiler gave warnings about this, since after refactoring, some language extensions might not be needed anymore, and hence should be removed since fewer language extensions mean more stable and portable code no? What do you think?

5 7

Haskell SDL
by michael millard 12 Mar '09

12 Mar '09

hi, im having trouble setting up the haskell sdl bindings for windows vista im trying to install using minGW but its just not installing correctly. does anyone know any good tutorials for this? _________________________________________________________________ Get 30 Free Emoticons for your Windows Live Messenger http://www.livemessenger-emoticons.com/funfamily/en-ie/

1 0

How to get program command line arguments in Unicode-aware way (Unix/Linux)?
by Dimitry Golubovsky 12 Mar '09

12 Mar '09

Hi, I am trying to process command line arguments that may contain Unicode (cyrillic in this example) characters. The standard GHC's getArgs seems to pass whatever was obtained from the underlying C library without any regard to encoding, e. g the following program (testarg.hs): module Main where import System.Environment main = do x <- getArgs mapM (putStrLn . show) x being invoked (ghc 6.10.1) runghc testarg -T 'при<в>ет' prints the following: "-T" "\208\191\209\128\208\184<\208\178>\208\181\209\130" (not correct, all bytes were passed without proper encoding) Is there any way to get program arguments in GHC Unicode-aware? Or at least assuming that they are always in UTF-8? Something like System.IO.UTF8, but for command line arguments? Thanks. PS: BTW runhugs testarg -T 'при<в>ет' prints: "-T" "\1087\1088\1080<\1074>\1077\1090" which is correct. -- Dimitry Golubovsky Anywhere on the Web

3 2

Loading 3D points & normals into OpenGL?
by Duane Johnson 12 Mar '09

12 Mar '09

Hi, I am considering writing a VRML (.wrl) parser so that I can load points and normals for a game I'm making in Haskell. Is there something around that will already do the trick? Or perhaps another format is preferred and already supported? Thanks, Duane Johnson (canadaduane) http://blog.inquirylabs.com/

4 4