[Haskell-cafe] ANNOUNCE: Haskell XML Toolbox Version 8.0.0
Haskell XML Toolbox 8.0.0 I would like to announce a new version of the Haskell XML Toolbox. The main change is the separation of the old filter interface into a separate package hxt-filter. This makes the API simpler and more readable. The filter API will not be developed further. But all the functionality ist still available. The filter package hxt-filter must be installed on top of hxt. The darcs repos for HXT are at http://darcs.fh-wedel.de/hxt http://darcs.fh-wedel.de/hxt-filter Furthermore there is a new wiki page http://www.haskell.org/haskellwiki/HXT/Conversion_of_Haskell_data_from/to_XML describing the serialization and deserialization of Haskell values to/from XML data. More information and download: http://www.fh-wedel.de/~si/HXmlToolbox/index.html There's also a copy on Hackage. Please email comments, bugs, etc. to [EMAIL PROTECTED] or [EMAIL PROTECTED] Uwe -- University of Applied Sciences, Wedel, Germany http://www.fh-wedel.de/~si/index.html ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] approximating pi
Benjamin L. Russell: Assuming the square had 100 pixels per side, on the average, approximately how many random pixels should be plotted in the square before obtaining a reasonably good estimate of pi? Nothing to do with Haskell... What do you mean by reasonable? This Monte-Carlo procedure is very inefficient anyway. The relative error falls as 1/sqrt(N) where N is the number of samples, so, several hundred thousands of samples may give you just three significant digits. And, at any rate, this has nothing to do with pixels, what, introduce one more source of errors through the truncation of real randoms? Jerzy Karczmarczuk ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Haskell as a general purpose programming language...
Hi all, just a thought: is it just me, or is Haskell really really really the coolest general-purpose programming language? If I had the time, I would love to do a re-implementation of Harpy [1] using category-extras [2]. Martin [1] http://hackage.haskell.org/cgi-bin/hackage-scripts/package/harpy-0.4 [2] http://hackage.haskell.org/cgi-bin/hackage-scripts/package/category-extras-0.2 signature.asc Description: OpenPGP digital signature ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Haskell as a general purpose programming language...
Hello Martin, Monday, April 28, 2008, 3:51:11 PM, you wrote: just a thought: is it just me, or is Haskell really really really the coolest general-purpose programming language? the only Haskell drawback is that it forces you to hate all other programming languages (c) people :) -- Best regards, Bulatmailto:[EMAIL PROTECTED] ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] FFI Query: How to free a CString allocated in Haskell
Hi, I am calling a Haskell function from my C code, wherein I export a Haskell function xyz that takes as an argument string and returns a string: Foreign export ccall xyz::CString - IO CString xyz = do -- extract the input string and act on it-- -- at the end I return a string like this newCString str Now once I call this function from C code, I am freeing the allocated memory using free function. I want to confirm that this is the right thing to do. - Anurag ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] FFI Query: How to free a CString allocated in Haskell
Hello Verma, Monday, April 28, 2008, 4:11:51 PM, you wrote: newCString str Now once I call this function from C code, I am freeing the allocated memory using free function. I want to confirm that this is the right thing to do. yes, i've traced this function down to mallocArray - mallocBytes - malloc() calls -- Best regards, Bulatmailto:[EMAIL PROTECTED] ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] question about GHC and Unicode
On Sun April 27 2008 2:02:25 pm Don Stewart wrote: zefria: In GHC there's a GHC.Unicode library, but for a string such as *AIOO, a GHC compiled program prints it as a string of unknown characters, and in the interpreter, the string evaluates to a string of escape sequences instead of displaying properly. Is there a way to get/activate unicode support in GHC? GHC supports unicode internally, and String and Char are all unicode. To do unicode IO however, you need to use the utf8-string package: http://hackage.haskell.org/cgi-bin/hackage-scripts/package/utf8-string Just import and use IO functions from System.IO.UTF8, and that's it! That's a wonderful interface, but unfortunately it appears to assume that your Unicode I/O is always UTF-8, and never UTF-16. I happen to deal with more UTF-16 data than UTF-8 over here at the moment. (Did I mention UTF-7 or UTF-EBCDIC? horrors...) There is a Haskell binding for iconv for those that need UTF-16, but it doesn't appear to have as convenient an interface. -- John ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] [help] Program immplementing an alarm
Antonio Regidor García wrote: I'm trying to implement an alarm in Haskell and wrote the following code: http://hpaste.org/7201 But it doesn't work as expected. Hi, The getLine function ties up stdin. So the system function isn't able to proceed until getLine completes, even when it is in another thread. Try using System.Process.runInteractiveProcess instead of system. Don't forget to call waitForProcess on the process handle when you are finished, or you will be left with a zombie. Hope this helps, Yitz ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] maintainer of StorableVector
I sent some patches to the maintainer of StorableVector, Spencer Janssen http://code.haskell.org/~sjanssen/storablevector but did not get some feedback and the patches are still not applied. Any idea what might went wrong? ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] approximating pi
On Mon, Apr 28, 2008 at 09:47:44AM +0200, [EMAIL PROTECTED] wrote: Benjamin L. Russell: Assuming the square had 100 pixels per side, on the average, approximately how many random pixels should be plotted in the square before obtaining a reasonably good estimate of pi? Nothing to do with Haskell... What do you mean by reasonable? This Monte-Carlo procedure is very inefficient anyway. The relative error falls as 1/sqrt(N) where N is the number of samples, so, several hundred thousands of samples may give you just three significant digits. And, at any rate, this has nothing to do with pixels, what, introduce one more source of errors through the truncation of real randoms? Indeed, Monte-Carlo is generally very inefficient, but it does have the advantage of often allowing one to write very simple code that is thus easily shown to be correct, and (as long as the random number generator is good!!!) to get rigorously-correct error bars, which is something that more efficient methods often struggle on. For simple tasks like computing pi or integrating smooth functions, this doesn't matter much, but it's quite a big deal when considering methods such as Quantum Monte Carlo (although, alas the fermion problem means that for most QMC calculations one *doesn't* get a rigorous error bar on the calculation... it's just better than almost any other method, and for bosons you *can* get a rigorous error bar). -- David Roundy Department of Physics Oregon State University ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Re: Caching the Result of a Transaction?
The garbage collector never gets to collect either the action used to populate the cached value, or the private TMVar used to hold the cached value. A better type for TIVal is given below. It is a newtype of a TVal. The contents are either a delayed computation or the previously forced value. Thew newTIVal(IO) functions immediately specify the delayed action. The newEmptyTIVal(IO) functions create a private TMVar that allows the delayed action to be specified once later. Note the use of tryPutTMVar to return a Bool instead of failing, in the event that the user tries to store more that one action. When force is called, the previous action (and any private TMVar) are forgotten. The garbage collector might then be free to collect them. -- Chris -- By Chris Kuklewicz (April 2008), public domain module TIVal(TIVal,newTIVal,newTIValIO,force,cached) where import Control.Applicative(Applicative(..)) import Control.Concurrent.STM(STM,TVar,newTVar,newTVarIO,readTVar,writeTVar ,TMVar,newEmptyTMVar,newEmptyTMVarIO,tryPutTMVar,readTMVar) import Control.Monad(Monad(..),join,liftM2) import System.IO.Unsafe(unsafePerformIO) newtype TIVal a = TIVal (TVar (Either (STM a) a)) -- the non-empty versions take a computation to delay newTIVal :: STM a - STM (TIVal a) newTIVal = fmap TIVal . newTVar . Left newTIValIO :: STM a - IO (TIVal a) newTIValIO = fmap TIVal . newTVarIO . Left -- The empty versions stage things with a TMVar, note the use of join -- Plain values 'a' can be stored with (return a) newEmptyTIVal :: STM ( TIVal a, STM a - STM Bool) newEmptyTIVal = do private - newEmptyTMVar tv - newTVar (Left (join $ readTMVar private)) return (TIVal tv, tryPutTMVar private) newEmptyTIValIO :: IO ( TIVal a, STM a - STM Bool ) newEmptyTIValIO = do private - newEmptyTMVarIO tv - newTVarIO (Left (join $ readTMVar private)) return (TIVal tv, tryPutTMVar private) -- force will clearly let go of the computation (and any private TMVar) force :: TIVal a - STM a force (TIVal tv) = do v - readTVar tv case v of Right a - return a Left wait - do a - wait writeTVar tv (Right a) return a -- Conal's cached function. This is actually safe. cached :: STM a - TIVal a cached = unsafePerformIO . newTIValIO -- The instances instance Functor TIVal where f `fmap` tiv = cached (f `fmap` force tiv) instance Applicative TIVal where pure x = cached (pure x) ivf * ivx = cached (force ivf * force ivx) instance Monad TIVal where return x = cached (return x) tiv = k = cached (force tiv = force . k) instance Applicative STM where pure x = return x ivf * ivx = liftM2 ($) ivf ivx ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] maintainer of StorableVector
lemming: I sent some patches to the maintainer of StorableVector, Spencer Janssen http://code.haskell.org/~sjanssen/storablevector but did not get some feedback and the patches are still not applied. Any idea what might went wrong? I think its just unmaintained. -- Don ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] CPU simulation library?
I want to simulate various process/thread scheduling algorithms for a small project. I'm trying to compare the performance versus power consumption of several algorithms. If anyone knows of some Haskell code along those lines I'd love to hear about it. Thanks in advance! Justin p.s. I can find non-Haskell code on my own - I'm hoping there is something in my favorite language, but any particularly good suggestions are also welcome. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Simplest possible Fasta shootout entry. How do I zap the ugly line? Suggest any other improvements.
r.kelsall: (Extracting these questions from my previous thread for clarity.) Below is my simplest possible program to solve the Fasta shootout benchmark. http://shootout.alioth.debian.org/gp4/benchmark.php?test=fastalang=all http://haskell.org/haskellwiki/Shootout/Fasta I can see one remaining flaw - the line marked 'Ugly'. What's the best way to get rid of this line? Any other suggestions for simplifying or improving the program would also be interesting. This code is about three or four times slower that the current fastest GHC entry for the Fasta benchmark. I'll elaborate it for speed when I've produced the best version regardless of speed. This is quite nice, and you can probably match the current entry by switching to lazy ByteString IO, as the current entry does. -- Don ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Simplest possible Fasta shootout entry. How do I zap the ugly line? Suggest any other improvements.
Don Stewart wrote: r.kelsall: (Extracting these questions from my previous thread for clarity.) Below is my simplest possible program to solve the Fasta shootout benchmark. http://shootout.alioth.debian.org/gp4/benchmark.php?test=fastalang=all http://haskell.org/haskellwiki/Shootout/Fasta I can see one remaining flaw - the line marked 'Ugly'. What's the best way to get rid of this line? Any other suggestions for simplifying or improving the program would also be interesting. This code is about three or four times slower that the current fastest GHC entry for the Fasta benchmark. I'll elaborate it for speed when I've produced the best version regardless of speed. This is quite nice, and you can probably match the current entry by switching to lazy ByteString IO, as the current entry does. -- Don Thanks Don :) I'll try that. The thing I really like about this version is that it localizes the 'breaking the lines at 60 characters' part of the program to just one function. I would never have thought to do this in a language other than Haskell and looking through most of the other language submissions for Fasta I can't see any that abstract this feature. I seem to be able to think more clearly in Haskell. Richard. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Trouble compiling collections-0.3 (from Hackage)
ccing Haskell Cafe in case anyone else is interested in my answer.. Bryan O'Sullivan wrote: Adrian Hey wrote: I don't think anyone is interested in working on this or maintaining it, so it's probably best not to use it for new stuff. If nobody has stepped up yet, I'll take it over. It would be a shame to see it bitrot. I think the main problem is that the collections package contained a number of different sub-projects that should really be separated out (and indeed have been to some extent), as they were progressing at rather different rates and it was never clear when would be a good time to do a new release of collections as a whole, so it was never done. There's the abstract collections class API's which were really what Jean-Philippe Bernardy was working on. I'm not sure if he's still interested in that so you'd better ask him. I think he's more interested in Yi these days :-) Then there's the AVL tree lib which has been put in hackage as a separate package already. I kinda washed my hands of this too, but I guess I'd better change my mind about that as if I don't maintain it I don't think any one else will. Then there's the relatively trivial AVL based Data.Set/Map clones which aren't in hackage yet but I guess could be put there without too much work. Then there's all the generalised trie stuff which has stalled for the time being, at least until type families are available. You'll probably see some more action on this in the summer as it's the subject of Jamie Brandons GSoC project. Regards -- Adrian Hey ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Re: Caching the Result of a Transaction?
Hi Chris,
Thanks a bunch for the new angle.
Question comments:
* I like the simplicity of using a single TVar whose state reflects the
not-computed/computed state of the IVal.
* I also like the public interface of taking an STM argument (newTIVal(IO))
over returning a sink (newEmptyTIVal(IO)), which came from some non-STM
thinking. In fact, maybe 'cached' is a better public interface yet. I'm
going to try it out, renaming cached to ival. (Oh yeah, I'm shortening
TIVal to IVal.)
* Why tryPutTMVar in place of putTMVar? Perhaps to encourage checking that
var hasn't been written?
* A perhaps prettier version of force:
force (TIVal tv) = readTVar tv = either compute return
where
compute wait = do a - wait
writeTVar tv (Right a)
return a
* The Applicative STM instance can be simplified:
instance Applicative STM where { pure = return; (*) = ap }
Cheers, - Conal
On Mon, Apr 28, 2008 at 7:40 AM, ChrisK [EMAIL PROTECTED]
wrote:
The garbage collector never gets to collect either the action used to
populate the cached value, or the private TMVar used to hold the cached
value.
A better type for TIVal is given below. It is a newtype of a TVal. The
contents are either a delayed computation or the previously forced value.
Thew newTIVal(IO) functions immediately specify the delayed action.
The newEmptyTIVal(IO) functions create a private TMVar that allows the
delayed action to be specified once later. Note the use of tryPutTMVar to
return a Bool instead of failing, in the event that the user tries to store
more that one action.
When force is called, the previous action (and any private TMVar) are
forgotten. The garbage collector might then be free to collect them.
--
Chris
-- By Chris Kuklewicz (April 2008), public domain
module TIVal(TIVal,newTIVal,newTIValIO,force,cached) where
import Control.Applicative(Applicative(..))
import
Control.Concurrent.STM(STM,TVar,newTVar,newTVarIO,readTVar,writeTVar
,TMVar,newEmptyTMVar,newEmptyTMVarIO,tryPutTMVar,readTMVar)
import Control.Monad(Monad(..),join,liftM2)
import System.IO.Unsafe(unsafePerformIO)
newtype TIVal a = TIVal (TVar (Either (STM a) a))
-- the non-empty versions take a computation to delay
newTIVal :: STM a - STM (TIVal a)
newTIVal = fmap TIVal . newTVar . Left
newTIValIO :: STM a - IO (TIVal a)
newTIValIO = fmap TIVal . newTVarIO . Left
-- The empty versions stage things with a TMVar, note the use of join
-- Plain values 'a' can be stored with (return a)
newEmptyTIVal :: STM ( TIVal a, STM a - STM Bool)
newEmptyTIVal = do
private - newEmptyTMVar
tv - newTVar (Left (join $ readTMVar private))
return (TIVal tv, tryPutTMVar private)
newEmptyTIValIO :: IO ( TIVal a, STM a - STM Bool )
newEmptyTIValIO = do
private - newEmptyTMVarIO
tv - newTVarIO (Left (join $ readTMVar private))
return (TIVal tv, tryPutTMVar private)
-- force will clearly let go of the computation (and any private TMVar)
force :: TIVal a - STM a
force (TIVal tv) = do
v - readTVar tv
case v of
Right a - return a
Left wait - do a - wait
writeTVar tv (Right a)
return a
-- Conal's cached function. This is actually safe.
cached :: STM a - TIVal a
cached = unsafePerformIO . newTIValIO
-- The instances
instance Functor TIVal where
f `fmap` tiv = cached (f `fmap` force tiv)
instance Applicative TIVal where
pure x = cached (pure x)
ivf * ivx = cached (force ivf * force ivx)
instance Monad TIVal where
return x = cached (return x)
tiv = k = cached (force tiv = force . k)
instance Applicative STM where
pure x = return x
ivf * ivx = liftM2 ($) ivf ivx
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Confused by instances
Hello, Haskellers, I feel like I'm missing something obvious, but here's some example code: module Instance where data Value = Value Integer class ValueClass a where fromValue :: Value - a instance ValueClass Bool where fromValue (Value n) = n /= 0 instance ValueClass String where fromValue (Value n) = show n instance (Num a) = ValueClass a where fromValue (Value n) = fromInteger n The Bool instance compiles fine. The String instance fails, and that would be because String is a synonym for [Char], and we can't create instances for those. There's a slight hack in the Prelude which allows you to define your own list show function, and that's exactly what gets done by Show Char, right? I'm sorry about calling it a hack but, well, it is. :) What I'm really confused by is the response to instance (Num a) = ValueClass a -- what I am trying to say is if a is an instance of Num, then can be an instance of ValueClass too, and here's how. But ghc says: Instance.hs:14:0: Illegal instance declaration for `ValueClass a' (All instance types must be of the form (T a1 ... an) where a1 ... an are distinct type *variables* Use -XFlexibleInstances if you want to disable this.) In the instance declaration for `ValueClass a' I don't understand why what I'm doing above is different from, for example, instance Show a = Show [a], or Monad m = MonadState s (StateT s m) ... I imagine that it's related to the fact that the difference between those two and what I have is that they have either the same class on each side of the =, or new type variables on the right. But I'm having trouble abstracting from that to the general rule. Any help would be greatly appreciated. In context, I'm attempting to wrap up different types inside a data type, and rather than extracting the value with getBoolean, getInteger, getString etc, I thought I would let type classes work for me, and just say fromValue instead. Is this silly? You can tell me, I won't be hurt. cheers, Fraser. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Confused by instances
On Apr 28, 2008, at 16:22 , Fraser Wilson wrote: instance (Num a) = ValueClass a where fromValue (Value n) = fromInteger n What I'm really confused by is the response to instance (Num a) = ValueClass a -- what I am trying to say is if a is an instance of Num, then can be an instance of ValueClass too, and here's how. The format is instance [context =] classname instance. Your classname is ValueClass. Your instance is a. a is not of the form (T a1 ... an). (How to fix it? Not sure, and am trying to get myself out of here and on the road :) -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] [EMAIL PROTECTED] system administrator [openafs,heimdal,too many hats] [EMAIL PROTECTED] electrical and computer engineering, carnegie mellon universityKF8NH ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Confused by instances
On Mon, Apr 28, 2008 at 10:50 PM, Brandon S. Allbery KF8NH [EMAIL PROTECTED] wrote: The format is instance [context =] classname instance. Your classname is ValueClass. Your instance is a. a is not of the form (T a1 ... an). But neither is instance (Show a) = Show [a] ... *sound of penny dropping * Ah. Wood, trees, all that. Thanks! cheers, Fraser. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] pls help about subtree
Assume a tree is a subtree of the other if all elements of the first tree is included in the second with the exact structure; all parent-child relations are preserved with their order. data Tree = Empty | Leaf Int | Node (Int,Tree,Tree) subtree:: Tree - Tree - Bool How can i start to write this code with this data structure? Example; tree1 tree2 3 / \ 5 6 / \ 7 4 13 /\ 3 12 / \ / \ 5 6 9 6 / \ / \ 7 4 11 7 / / \ 2 3 4 Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Confused by instances
2008/4/28 Fraser Wilson [EMAIL PROTECTED]:
On Mon, Apr 28, 2008 at 10:50 PM, Brandon S. Allbery KF8NH
[EMAIL PROTECTED] wrote:
The format is instance [context =] classname instance.
Your classname is ValueClass.
Your instance is a. a is not of the form (T a1 ... an).
But neither is
instance (Show a) = Show [a] ...
Yes it is, it's just a weird looking T, (namely []). This works just as well:
instance (Show a) = Show ([] a) ...
(Unless that's not H98, but I think it is)
Instances have to have concrete constructors at their heads is for
technical reasons, namely you can't in general do type inference with
unrestricted instances. It is a pattern-matching algorithm as you'd
expect, but it goes *backwards*, from right to left; i.e it sees the
pattern [a] and generates a new constraint Show a. This is contrary to
the intuition that it has a big set of instances, and when it sees,
say, Show Int it adds Show [Int]. The direction of the arrow can be
misleading :-)
To answer your other question, no, there is no list show hack. What
is being complained about is that you're using a type synonym as an
instance. If you just expand the synonym everything works fine. This
is not specific to lists, nor show. You're just not allowed to use
synonyms in instances (and as the compiler suggests, this restriction
can be lifted with {-# LANGUAGE FlexibleInstances #-}).
Luke
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Confused by instances
2008/4/28 Fraser Wilson [EMAIL PROTECTED]:
what I am trying to say is if a is an instance of Num, then
can be an instance of ValueClass too, and here's how.
Oh, didn't answer this one. This is almost canned response, questions
like this get asked weekly on this list.
Short answer: you can't.
Longer answer: you can, but you have to wrap it in a newtype, which is
irritating.
newtype NumValue a = NumV a
instance (Num a) = ValueClass (NumV a) where
fromValue (Value n) = NumV (fromInteger n)
Essentially you have to tell the compiler when you use this
instance. So you still get all the power, but with less convenience
(than the impossible thing you want).
Abridged longest answer: you can, and you don't need a newtype, but
only if you're The Devil. Here's an explanation, but *please do not
do this*! It's unpredictable, poor style, a bad habit, nonmodular,
etc. etc. etc.
First enable undecidable instances:
{-# LANGUAGE UndecidableInstances #-}
With this pragma you are forfeiting your right to a terminating
compiler. The compiler may instance stack overflow or run forever
for no discernible reason.
Now you are allowed exactly one instance of the form you desire:
instance (Num a) = ValueClass a where
fromValue (Value n) = fromInteger n
If you're lucky, you might be able to define some well-formed
instances in addition and have everything behave. It *will* break if
you add another such instance, for example:
instance (Read a) = ValueClass a where ...
Because when the compiler sees fromValue, it will try to match it
against the head of an instance. Both the Num and the Read forms
match every type, so it will *pick one arbitrarily*, without
backtracking. So if you wanted the Read one and it picked the Num
one, you are permanently out of luck and you basically have to scrap
everything.
So, yeah, there's a little excursion into the dirty corners of the
typeclass system. If you don't want to get spontaneously eaten by a
bear, use a newtype as above :-).
And now it's time to go make/edit a wiki page on the subject.
Luke
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Confused by instances
On Mon, Apr 28, 2008 at 11:33 PM, Luke Palmer [EMAIL PROTECTED] wrote: To answer your other question, no, there is no list show hack. Perhaps hack was a strong word. I'm not referring to type synonyms, but to the fact that Prelude's show class happens to have a special show function for lists, which happens to be handy when writing an instance for Show Char. I find the coupling here (between Show and a particular instance of Show, namely Show Char) to be disturbing -- Show should not (if you ask me) contain special machinery for specific instances. Not that I'll lose sleep over it, and of course the benefit (having pretty strings) far outweighs the philosophical cost. The showList function has always struck me as a bit of a wart though. cheers, Fraser. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Confused by instances
On Mon, Apr 28, 2008 at 3:47 PM, Fraser Wilson [EMAIL PROTECTED] wrote: On Mon, Apr 28, 2008 at 11:33 PM, Luke Palmer [EMAIL PROTECTED] wrote: To answer your other question, no, there is no list show hack. Perhaps hack was a strong word. I'm not referring to type synonyms, but to the fact that Prelude's show class happens to have a special show function for lists, which happens to be handy when writing an instance for Show Char. Oh, showList, that hack! Right, I had forgotten about that :-) Luke ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] pls help about subtree
2008/4/28 cetin tozkoparan [EMAIL PROTECTED]: Assume a tree is a subtree of the other if all elements of the first tree is included in the second with the exact structure; all parent-child relations are preserved with their order. data Tree = Empty | Leaf Int | Node (Int,Tree,Tree) Bit of a nitpick: Haskell folks would usually leave off the tuple on the Node constructor, since constructors tuple for you already. So Tree would become: data Tree = Empty | Leaf Int | Node Int Tree Tree But it's (almost) equivalent either way. subtree:: Tree - Tree - Bool How can i start to write this code with this data structure? The way most tree algorithms are done is by recursion. Maybe it would help if you saw the problem in a more explicit form: A tree t is a subtree of a tree u if: t = u, u is a Node and t is a subtree of the left child, or u is a Node and t is a subtree of the right child. That definition should roughly translate into an inefficient implementation. There's probably a more efficient one somewhere, but it's somewhere other than the top of my head. Luke ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Confused by instances
On Mon, Apr 28, 2008 at 11:46 PM, Luke Palmer [EMAIL PROTECTED] wrote: 2008/4/28 Fraser Wilson [EMAIL PROTECTED]: what I am trying to say is if a is an instance of Num, then can be an instance of ValueClass too, and here's how. Oh, didn't answer this one. This is almost canned response, questions like this get asked weekly on this list. Thanks for the detailed response. In the end, I've just selected the types I'm really interested in, and added instances for those. It works a treat. cheers, Fraser. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Something like optimistic evaluation
Hello Somewhat on the topic of optimistic evaluation, I've just thought of another way to evaluate thunks. When the program is about to block on some IO, what if we start a thread to evaluate (any) unevaluated thunks. We'll have additional system thread, but the blocked one will not actually consume any processor time. This would take place only when the program is compiled as threaded and run with -N k, k1. The RTS knows at least about some operations that will block, those which IO operations are implemented with. for example. It could merely start a process of evaluating any (or something more clever) outstanding thunks right before going into one of those operations and stop it when it's back. Of course, it's not like optimistic evaluation because we don't avoid creating thunks. But in a sense it's similar. It could also be compared with incremental garbage collection :) Has something like that been done, discussed? ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Something like optimistic evaluation
On Apr 28, 2008, at 5:09 PM, Daniil Elovkov wrote: Somewhat on the topic of optimistic evaluation, I've just thought of another way to evaluate thunks. When the program is about to block on some IO, what if we start a thread to evaluate (any) unevaluated thunks. We'll have additional system thread, but the blocked one will not actually consume any processor time. Seems like this would be bad in cases where the programmer is trying to exploit laziness, such as using infinite lists, no? - Jake ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Confused by instances
On Apr 28, 2008, at 17:47 , Fraser Wilson wrote: Perhaps hack was a strong word. I'm not referring to type synonyms, but to the fact that Prelude's show class happens to have a special show function for lists, which happens to be handy when writing an instance for Show Char. I find the coupling here (between Show and a particular instance of Show, namely Show Char) to be disturbing -- Show should not (if you ask me) contain special machinery for specific instances. Not that I'll lose sleep over it, and of course the benefit (having pretty strings) far outweighs the philosophical cost. This is what comes of overloading lists as strings. (And what goes along with it, namely slowness. See Data.ByteString for a solution that solves both problems but sadly is unlikely to replace String) -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] [EMAIL PROTECTED] system administrator [openafs,heimdal,too many hats] [EMAIL PROTECTED] electrical and computer engineering, carnegie mellon universityKF8NH ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Glome.hs-0.3 (bounding objects, heterogeneous lists)
Andrew Coppin wrote: Well, for example, Haskell doesn't have hetrogenous lists - which are trivial in any OOP language. That's quite awkward to get around. Also, both spheres and cylinders have a radius property, but then you end up with name clashes. Again, a non-issue in OOP languages. [I gather there's some new GHC language extension to partially solve this one - but only when types are statically known. In the general case, you'd have to write a radius class and define instances... yuck!] It's funny you mention that, those are actually problems I ran into, but (having learned my lesson the hard way in Ocaml), I decided not to try and force the language to do things my way, but instead try to do things in a way that Haskell makes easy. For instance, I started out by making each primitive a separate type (Sphere, Triangle, etc...), and then made a type class that defines a ray intersection method (and a few other odds and ends), but then I realized that I can't do something as simple as store a list of primitives (or if there is in fact a way, I'm not aware of it). Instead, I made a single large sum type with all my primitives: so that now I can easily make a list of primitives. (In fact, that's what a Group is.) I also gave up on using named fields, since coming up with a separate name for each one gets ugly: instead of radius, you have sphere_radius, cylinder_radius, cone_radius disc_radius, etc... All of which works, but now it's a pain to add new primitives. And *all* supported primitives must be defined in a single monolithic module. And you can't reuse spheres as bounding volumes without either reimplementing them or loosing type safety. The part about spheres and bounding spheres is actually not so much of a problem: I implemented a general Bound primitive that can bound any primitive with any other. In general, you would use a simple object like a sphere or box as the left argument and a complex object as the right argument. data Solid = ... | Bound Solid Solid ... and then to intersect the Bound object, you first do a shadow test on the left object before testing the right object: rayint :: Solid - Ray - Flt - Texture - Rayint rayint (Bound sa sb) r d t = let (Ray orig _) = r in if inside sa orig || shadow sa r d then rayint sb r d t else RayMiss Some kinds of primitives aren't likely to work well for bounding since they don't have a well-defined inside and outside (like triangles and discs), but I'd rather provide maximum flexibility and assume that users know how to use it sensibly. http://www.haskell.org/haskellwiki/Glome_tutorial#Bounding_Objects As for the maintenance issues, that is still a problem. It would be nice to split all the individual primitives into their own modules. Sebastian Sylvan wrote: On 4/27/08, Jim Snow [EMAIL PROTECTED] wrote: For instance, I started out by making each primitive a separate type (Sphere, Triangle, etc...), and then made a type class that defines a ray intersection method (and a few other odds and ends), but then I realized that I can't do something as simple as store a list of primitives (or if there is in fact a way, I'm not aware of it). You can, by using a wrapper type which wraps up any instance of the Intersect class: data Intersectable = forall a. Intersect a = MkIntersectable a For convenience you probably want to instantiate this wrapper in the class itself: instance Intersect Intersectable where rayIntersection (MkIntersectable x) ray = rayIntersection x ray boundingVolume (MkIntersectable x) = boundingVolume x -- etc... Now you can stick Intersectables in lists etc. I think that sounds like what I ought to be doing. -jim ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Re: Caching the Result of a Transaction?
The problem I have with all of these STM-based solutions to this problem is that they don't actually cache until the action fully executes successfully. For example, if you have a :: TIVal a, and f :: a - TIVal b, and you execute force (a = f) and the action returned by f executes retry for whatever reason, then the caching done in a gets undone. Ideally I want to be able to provide some proof that the result of a is pure and have it committed immediately when it finishes. Every attempt I've had so far to solve this problem ends up being some type of the form newtype X a = IO (STM (Either a (X a))) which has its own problems. -- ryan ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] HaXml and the XHTML 1.0 Strict DTD
Is anyone using HaXml to validate XHTML Strict? The old 1.13.2 version has some bugs in how it handles attributes that stop me from using it. It handled the DTD parsing fine. The most-recent darcs version relies on a newer ByteString than I have, so it is not easy for me to test it. A recent (this year) darcs version failed to parse the DTD, yielding this error: validate: In a sequence: in content spec of ELEMENT decl: head When looking for a non-empty sequence with separators: In a sequence: Expected % but found | in file xhtml1 at line 252 col 50 when looking for a content particle when looking for a content particle This is the context: !-- Document Head ===-- !ENTITY % head.misc (script|style|meta|link|object)* !-- content model is %head.misc; combined with a single title and an optional base element in any order -- !ELEMENT head (%head.misc;, ((title, %head.misc;, (base, %head.misc;)?) | (base, %head.misc;, (title, %head.misc; I appreciate any advice on this. cheers peter ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] FFI Query: How to free a CString allocated in Haskell
On 28/04/2008, at 7:23 PM, Bulat Ziganshin wrote: Hello Verma, Monday, April 28, 2008, 4:11:51 PM, you wrote: newCString str Now once I call this function from C code, I am freeing the allocated memory using free function. I want to confirm that this is the right thing to do. yes, i've traced this function down to mallocArray - mallocBytes - malloc() calls Depending on how you feel about it, Bulat's response either entirely answers your question or is neither here nor there. If you want to write portable code, you should check with the FFI spec. If you just want to write code that works with GHC as it presently is, you can strace it or whatever. As Bulat has already given you the pragmatic answer, allow me to grind my axe a little. In the FFI spec on p33, we find: newCString :: String - IO CString newCStringLen :: String - IO CStringLen Allocate a memory area for a Haskell string and marshal the string into its C representation. There are two variants of the routine, one for each supported string representation. The memory area allocated by these routines may be deallocated using MarshalAlloc.free. OK, so back on p19 we find: free :: Ptr a - IO () Free a block of memory that was allocated with malloc, mallocBytes, realloc, reallocBytes, or any of the allocation functions from MarshalArray (see Section 5.9). and we can draw the inference (reading between the lines) that this function need not have a relation to C's malloc. (Roughly, the intention is to allow a Haskell implementation to use whatever memory management it likes.) So no, using C's free is not the right thing to do. It will probably work in the real world, but you should try to use Haskell's free if possible. cheers peter ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Re: Caching the Result of a Transaction?
On Apr 28, 2008, at 10:01 PM, Ryan Ingram wrote: [...] if you have a :: TIVal a, and f :: a - TIVal b, and you execute force (a = f) and the action returned by f executes retry for whatever reason, then the caching done in a gets undone. Dangit, you're right. You just rained on the parade! Hmm... Perhaps we could use the data structure Chris proposed and mix it with my first approach involving forkIO. Another thread can perform the caching in a separate transaction _or_ the main thread can perform the caching itself in case the extra thread doesn't get a chance first. This would ensure that it manages to get cached _sometime_ even when the main transaction itself retries, but it loses some of the elegance we had gained by not forking a new thread and still doesn't _guarantee_ that the computations are run only once. This is not ideal. :( - Jake ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Re: Caching the Result of a Transaction?
Thanks, Ryan, for the reminder and explanation of this problem. - Conal On Mon, Apr 28, 2008 at 8:01 PM, Ryan Ingram [EMAIL PROTECTED] wrote: The problem I have with all of these STM-based solutions to this problem is that they don't actually cache until the action fully executes successfully. For example, if you have a :: TIVal a, and f :: a - TIVal b, and you execute force (a = f) and the action returned by f executes retry for whatever reason, then the caching done in a gets undone. Ideally I want to be able to provide some proof that the result of a is pure and have it committed immediately when it finishes. Every attempt I've had so far to solve this problem ends up being some type of the form newtype X a = IO (STM (Either a (X a))) which has its own problems. -- ryan ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
RE: [Haskell-cafe] FFI Query: How to free a CString allocated in Haskell
Peter, A naïve question I have now after reading your mail. How do I call MarshallAlloc.free from my C code because that's where I need to free it? Anurag So no, using C's free is not the right thing to do. It will probably work in the real world, but you should try to use Haskell's free if possible. cheers peter ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] FFI Query: How to free a CString allocated in Haskell
On Apr 29, 2008, at 1:45 , Verma Anurag-VNF673 wrote: A naïve question I have now after reading your mail. How do I call MarshallAlloc.free from my C code because that's where I need to free it? Provide a Haskell wrapper function to deallocate / free it, as a parallel to the function that allocates and returns it to the C code. -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] [EMAIL PROTECTED] system administrator [openafs,heimdal,too many hats] [EMAIL PROTECTED] electrical and computer engineering, carnegie mellon universityKF8NH ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
