Re: [Haskell-cafe] ghc 7.2.1 and super simple DPH
Peter Braam wrote:
Hi -
I'm trying to compile DotP.hs from
http://www.haskell.org/haskellwiki/GHC/Data_Parallel_Haskell#A_simple_example
(see
below)
The compiler complains and says (twice in fact):
DotP.hs:17:33: Not in scope: `fromPArrayP'
Could someone help me out please? Thanks a lot!
The code you posted had some wrapping issues and was missing an
import. I've included a version of the code that compiles for
me here (using ghc fromgit HEAD) using:
ghc -c -Odph -fdph-par DotP.hs
HTH,
Erik
{-# LANGUAGE ParallelArrays #-}
{-# OPTIONS_GHC -fvectorise #-}
module DotP (dotp_wrapper) where
import qualified Prelude
import Data.Array.Parallel
import Data.Array.Parallel.Prelude
import Data.Array.Parallel.Prelude.Double
dotp_double :: [:Double:] - [:Double:] - Double
dotp_double xs ys = sumP [:x * y | x - xs | y - ys:]
dotp_wrapper :: PArray Double - PArray Double - Double
{-# NOINLINE dotp_wrapper #-}
dotp_wrapper v w = dotp_double (fromPArrayP v) (fromPArrayP w)
--
--
Erik de Castro Lopo
http://www.mega-nerd.com/
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Is it possible to represent such polymorphism?
Quoting Richard O'Keefe o...@cs.otago.ac.nz:
On 3/10/2011, at 7:15 AM, Du Xi wrote:
I guess this is what I want, thank you all. Although I still wonder
why something so simple in C++ is actually more verbose and
requires less known features in Haskell...What was the design
intent to disallow simple overloading?
It's not SIMPLE overloading you are asking for,
but AD HOC overloading, which may look simple, but really isn't.
Taking your C++ f() example, in what sense are the two functions
_the same function_?
I understand that functions with the same name but different type
signatures are not the same function, but are a family of
functions,probably representing the same concept.
Also
identityInHaskell::a-a
identityInHaskell x = x
--I don't know if this is implemented as runtime polymorphism or
compile-time polymorphism,
--but it is possible to implement this as compile-time polymorphism.
and
template typename T T IdentityInCPP( T x ){ return x; }
are families of different functions.
I think the problem I have encountered can be broken down to 2 problems:
1. Haskell doesn't have a way to represent mapping from one type to
another (consider metafunctions in C++), which is addressed by
TypeFamilies.
2. Haskell disallows ad-hoc overloading. But I think implementing
ad-hoc overloading itself should be no more complex than implementing
type classes, perhaps it would tear a hole somewhere else?
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ghc 7.2.1 and super simple DPH
Erik de Castro Lopo wrote: The code you posted had some wrapping issues and was missing an import. I should have also mentioned how I figured out what the missing import was. Firstly, I tried hoogle [0] but couldn't find it. I then realised that it must be part of DPH and that I had a copy of the DPH sources on my machine. Going to the DPH source tree I did: find . -name \*.hs | xargs grep ^fromPArrayP which showed up this: dph-common/Data/Array/Parallel.hs:fromPArrayP :: PArray a - [:a:] Cheers, Erik [0] http://www.haskell.org/hoogle/ -- -- Erik de Castro Lopo http://www.mega-nerd.com/ ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] How to compile this example code?
{- forgot to reply to list -}
This isn't Haskell syntax. Atleast not directly. It is either
hsc2hs[1] or c2hs [2]. Also see [3] for the difference between the
two. Soin order to compile that code you first have to run it through
aspecial preprocessor.
1 - http://www.haskell.org/ghc/docs/7.2.1/html/users_guide/hsc2hs.html2
- http://www.cse.unsw.edu.au/~chak/haskell/c2hs/3 -
http://stackoverflow.com/questions/6009031/difference-between-hsc2hs-and-c2hs
2011/10/3 Magicloud Magiclouds magicloud.magiclo...@gmail.com:
Hi,
I am learning to use data-flags, and got this example code:
import Data.Flags
newtype MyFlags = MyFlags CInt deriving (Eq, Flags)
#{enum MyFlags, MyFlags
, myFlag1 = C_FLAG1
, myFlag2 = C_FLAG2
, myFlag3 = C_FLAG3
}
I modified it trying to compile it. Well, I got illegal syntax at #{e.
In fact, I am not quite know the syntax here. Any clue?
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Is it possible to represent such polymorphism?
Although I still wonder why something so simple in C++ is actually more verbose and requires less known features in Haskell...What was the design intent to disallow simple overloading? The simple C++ overloading you want to add to Haskell, is in fact rather semantically complex, and it leads to undecidability of the type system. The inherent formal complexity here suggests that this form of overloading is highly unlikely to be the correct solution in practice to the problem you are trying to solve. And even if it were a technically correct solution, it is likely to be unmaintainable and fragile to code changes. There is a high probability that a more-formally-tractable solution exists, and that using it will improve your understanding of the problem at hand, and make your code more regular and robust to change. Regards, Malcolm ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Is it possible to represent such polymorphism?
Quoting Felipe Almeida Lessa felipe.le...@gmail.com:
On Sun, Oct 2, 2011 at 4:26 PM, Edward Z. Yang ezy...@mit.edu wrote:
What are you actually trying to do? This seems like a rather
unusual function.
If you're new to the language, most likely you're doing something
wrong if you need this kind of function. =)
--
Felipe.
{-# LANGUAGE TypeFamilies,FlexibleInstances #-}
module RicherListOp ( generalizedFilter,generalizedMap,generalizedFilterMap )
where
import Data.List
generalizedFilter pred = impl.expand3 where
impl (dL,dR,step) = generalizedFilterMap tf (dL+dR+1,step) where
tf s = if pred s then [s !! dL] else []
generalizedMap tf = generalizedFilterMap $ \x-[tf x]
generalizedFilterMap tf ns ls = impl {-$ expand2-} ns where
impl (len,step) = f ls where
f xs | length xs =len = (tf $ genericTake len xs) ++ (f $
genericDrop step xs)
f _ = []
class Expand3 t where
type Result3 t
expand3 :: t-Result3 t
instance (Integral a,Integral b)=Expand3 (a,b) where
type Result3 (a,b) = (a,b,Int)
expand3 (l,r) = (l,r,1)
instance (Integral a,Integral b,Integral c)=Expand3 (a,b,c) where
type Result3 (a,b,c) = (a,b,c)
expand3 = id
--instance (Integral a)=Expand3 a where
-- type Result3 a = (a,a,a)
-- expand3 r = (0,r,1)
--class Expand2 t where
-- type Result2 t
-- expand2 :: t-Result2 t
--instance (Integral a)=Expand2 (a,a) where
-- type Result2 (a,a) = (a,a)
-- expand2 = id
--instance (Integral a)=Expand2 a where
-- type Result2 a = (a,a)
-- expand2 a = (a,1)
examples:
generalizedFilterMap (\[x,y,z]- if(x==1z==1)then [y*10] else
[0]) (3,1) [1,2,3,4,1,2,1,3,1,4,1,5,2]
[0,0,0,0,20,0,30,0,40,0,0]
it :: [Integer]
generalizedFilter (\[x,y,z] - x==1z==1) (1,1) [1,2,3,4,1,2,1,3,1,4,1,5,2]
[2,3,4]
it :: [Integer]
The code commented out is what I still can't get working. (I'm no
longer trying to finish them. They are included just to illustrate my
idea). Of course, I could have simply used [Int] , (Num a)=[a] or
(Int,Int,Int), but I'm trying to write code as generic as possible.
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Is it possible to represent such polymorphism?
Quoting Andrew Coppin andrewcop...@btinternet.com: On 02/10/2011 07:15 PM, Du Xi wrote: In C++, the code is inferred from the types. (I.e., if a function is overloaded, the correct implementation is selected depending on the types of the arguments.) In Haskell, the types are inferred from the code. (Which is why type signatures are optional.) Really, it's just approaching the same problem from a different direction. Also, as others have said, you're probably just approaching the problem from the wrong angle. You don't design an object-oriented program the same way you'd design a procedural program; if you do, you end up with a horrible design. Similarly, you don't design a functional program the same way you would design an object-oriented one. It takes time (and experience) to figure out how to approach FP - or any other radically different paradigm, I suppose... ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe Quoting Brandon Allbery allber...@gmail.com: On Sun, Oct 2, 2011 at 15:17, sdiy...@sjtu.edu.cn wrote: Not for your meaning of clean. C++ is an object-oriented programming language; given a method call, it tries really hard to shoehorn the arguments to the call into some declared method somewhere along the inheritance chain. Haskell is a functional programming language; it is strongly typed, and typeclasses are a mechanism to allow that typing to be weakened in a strictly controlled fashion. In some sense, it's not *supposed* to be convenient, because the whole point is you're not supposed to throw arbitrarily-typed expressions at arbitrary functions. Instead, a properly designed program is characterized by its types; if the types are well designed for the problem being solved, they very nearly write the program by themselves. This doesn't mean that use of typeclasses / ad-hoc polymorphism is automatically a sign of a poor design, but it *does* mean you should think about what you're trying to do whenever you find yourself considering them. Nor does it mean that C++ is in some sense wrong; it means the languages are fundamentally different, and the appropriate design of a program is therefore also usually different between the two. -- brandon s allbery allber...@gmail.com wandering unix systems administrator (available) (412) 475-9364 vm/sms This has nothing to do with OOP or being imperative.It's just about types. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Is it possible to represent such polymorphism?
sdiy...@sjtu.edu.cn writes: This has nothing to do with OOP or being imperative. It's just about types. Of course, it's not necessarily linked to OOP, but OO languages - to the extent they have types - tend towards ad-hoc polymorphism instead of parametric polymorphism. There are different trade-offs, one is the lack of return-type overloading in C++. -k -- If I haven't seen further, it is by standing in the footprints of giants ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] ANN: OpenCL 1.0.1.3 package
Hello, all. I want to show you the OpenCL package. I have done this using Jeff Heard OpenCLRaw package, but I create a new one due the lack of updates of the former. # Where to get it * Hackage page (http://hackage.haskell.org/package/OpenCL) * Repository (https://github.com/zhensydow/opencl) * Bugs (https://github.com/zhensydow/opencl/issues) * Examples (https://github.com/zhensydow/opencl/tree/master/examples). # Things: * I write it's high-level binding to OpenCL libraries, but only because I added more types to hide most of the alloc/free of the API, and hide the enums using c2hs enums. * The worst problem of the OpenCLRaw is the bad types it use, I learn to fix 32/64 bits issues with c2hs, and test it on linux machines. * Tested on Linux + NVidia only. * Jason Dagit is helping with Windows, OSX testing in own fork, also the call-conv fork in github has changes to work on Windows Please, Consider it's on experimental status but it works, I need lots of feedbacks for detect posible errors, Thanks, ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ghc 7.2.1 and super simple DPH
Super helpful, thanks! Peter On Mon, Oct 3, 2011 at 12:54 AM, Erik de Castro Lopo mle...@mega-nerd.comwrote: Erik de Castro Lopo wrote: The code you posted had some wrapping issues and was missing an import. I should have also mentioned how I figured out what the missing import was. Firstly, I tried hoogle [0] but couldn't find it. I then realised that it must be part of DPH and that I had a copy of the DPH sources on my machine. Going to the DPH source tree I did: find . -name \*.hs | xargs grep ^fromPArrayP which showed up this: dph-common/Data/Array/Parallel.hs:fromPArrayP :: PArray a - [:a:] Cheers, Erik [0] http://www.haskell.org/hoogle/ -- -- Erik de Castro Lopo http://www.mega-nerd.com/ ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] HEADS-UP: security fix, please upgrade clientsession to = 0.7.3.1
Hello! Please be advised that clientsession 0.7.3.1 is vulnerable to timing attacks [1]. We have just released a fix and it's already on Hackage [2]. We advise all users of clientsession (and, consequently, Yesod) to upgrade as soon as possible to a version = 0.7.3.1. With a timing attack a malicious user may be able to construct a valid MAC for his message. However, the attacker is not able to recover the MAC key or the encryption key. So you don't need to change your keys, just upgrade ASAP. Cheers, =) [1] https://github.com/snoyberg/clientsession/pull/4 [2] http://hackage.haskell.org/package/clientsession-0.7.3.1 -- Felipe. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] haskell i18n best practices
Thanks for all the great information provided in this thread. The wiki page that Paulo originally linked had Vasyl's fantastic documentation for using his hgettext package, but it did not mention any of the other methods we discussed. I moved the gettext documentation to its own linked page and tried to collect together the general information from this thread. Please take a moment and look it over. Correct any mistakes I made. http://haskell.org/haskellwiki/Internationalization_of_Haskell_programs Rogan, especially, please look it over. I really had to read between the lines to come up with a clear and concise description of GF and what it does, so I may have gotten it wrong. Felipe, I put your wonderful example on its own linked page. Thanks, Yitz ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] HEADS-UP: security fix, please upgrade clientsession to = 0.7.3.1
On Mon, Oct 3, 2011 at 10:01 AM, Felipe Almeida Lessa felipe.le...@gmail.com wrote: With a timing attack a malicious user may be able to construct a valid MAC for his message. However, the attacker is not able to recover the MAC key or the encryption key. So you don't need to change your keys, just upgrade ASAP. If you are really paranoid, you may worry about a malicious user that created a valid cookie for an administrator expiring on 2030 while you still haven't upgraded. If have this level of security paranoia/consciousness, you may want to generate new keys. Just delete client_session_key.aes before restarting your application with the fixed clientsession = 0.7.3.1 and new, random keys will be generated for you. Cheers, =) -- Felipe. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] haskell i18n best practices
On Mon, Oct 3, 2011 at 10:05 AM, Yitzchak Gale g...@sefer.org wrote: Felipe, I put your wonderful example on its own linked page. Thanks, I'm always lazy with wikis =). I've corrected a few typos I've made on my e-mail there. Cheers, -- Felipe. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] How to compile this example code?
On Mon, Oct 3, 2011 at 3:50 PM, Roel van Dijk vandijk.r...@gmail.com wrote:
{- forgot to reply to list -}
This isn't Haskell syntax. Atleast not directly. It is either
hsc2hs[1] or c2hs [2]. Also see [3] for the difference between the
two. Soin order to compile that code you first have to run it through
aspecial preprocessor.
1 - http://www.haskell.org/ghc/docs/7.2.1/html/users_guide/hsc2hs.html2
- http://www.cse.unsw.edu.au/~chak/haskell/c2hs/3 -
http://stackoverflow.com/questions/6009031/difference-between-hsc2hs-and-c2hs
2011/10/3 Magicloud Magiclouds magicloud.magiclo...@gmail.com:
Hi,
I am learning to use data-flags, and got this example code:
import Data.Flags
newtype MyFlags = MyFlags CInt deriving (Eq, Flags)
#{enum MyFlags, MyFlags
, myFlag1 = C_FLAG1
, myFlag2 = C_FLAG2
, myFlag3 = C_FLAG3
}
I modified it trying to compile it. Well, I got illegal syntax at #{e.
In fact, I am not quite know the syntax here. Any clue?
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
I see. I was mislead, did not noticed the c binding part.
Thank you.
--
竹密岂妨流水过
山高哪阻野云飞
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Problem on using class as type.
Hi, I have a function: post :: (ToJson p, FromJson q) = String - String - String - Map.Map String p - IO q Now I'd like to call it like: r - post site token user.addMedia (Map.fromList [ (users, users :: ToJson) , (medias, medias :: ToJson) ]) So I got the problem. If I used things like users :: ToJson, then class used as a type error occurred. But if I did not use them, since users and medias were actually different types, then fromList failed, required the type of medias the same with users. How to resolve the conflict? -- 竹密岂妨流水过 山高哪阻野云飞 ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Problem on using class as type.
What about users :: ToJson a = a? On Tue, Oct 4, 2011 at 12:42 AM, Magicloud Magiclouds magicloud.magiclo...@gmail.com wrote: Hi, I have a function: post :: (ToJson p, FromJson q) = String - String - String - Map.Map String p - IO q Now I'd like to call it like: r - post site token user.addMedia (Map.fromList [ (users, users :: ToJson) , (medias, medias :: ToJson) ]) So I got the problem. If I used things like users :: ToJson, then class used as a type error occurred. But if I did not use them, since users and medias were actually different types, then fromList failed, required the type of medias the same with users. How to resolve the conflict? -- 竹密岂妨流水过 山高哪阻野云飞 ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Problem on using class as type.
On Mon, Oct 03, 2011 at 09:42:59PM +0800, Magicloud Magiclouds wrote: Hi, I have a function: post :: (ToJson p, FromJson q) = String - String - String - Map.Map String p - IO q Now I'd like to call it like: r - post site token user.addMedia (Map.fromList [ (users, users :: ToJson) , (medias, medias :: ToJson) ]) So I got the problem. If I used things like users :: ToJson, then class used as a type error occurred. But if I did not use them, since users and medias were actually different types, then fromList failed, required the type of medias the same with users. If users and medias are different types, then you cannot put them in the same list. The ToJson class provides a method toJson :: ToJson a = a - Json so presumably you need to call this method on users and medias, like ... (users, toJson users) , (medias, toJson medias) -Brent ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Problem on using class as type.
On 10/03/2011 10:42 PM, Magicloud Magiclouds wrote:
Hi,
I have a function:
post :: (ToJson p, FromJson q) = String - String - String -
Map.Map String p - IO q
Now I'd like to call it like:
r- post site token user.addMedia (Map.fromList [ (users, users :: ToJson)
, (medias, medias
:: ToJson) ])
So I got the problem. If I used things like users :: ToJson, then
class used as a type error occurred. But if I did not use them, since
users and medias were actually different types, then fromList failed,
required the type of medias the same with users.
How to resolve the conflict?
If 'users' and 'medias' are actually of a general type (like for all a
with ToJson a, users describes a value of type a), use Jesse's
suggestion. Otherwise (there is an a with ToJson a such that users
describes a value of type a), you might want to use existentials:
{-# LANGUAGE ExistentialQuantification #-}
data SomeToJson = forall a. (ToJson a) = SomeToJson a
instance ToJson SomeToJson where
toJson (SomeToJson x) = toJson x -- I guess your class looks like
this?
And then:
r - post site token user.addMedia $ Map.fromList
[(users, SomeToJson users), (medias, SomeToJson medias)]
As a last remark, I needed this pattern exactly once, namely for dealing
with rank 2 types in rendering functions using takusen. I can conclude
that requiring it is often an indicator for a major design flaw in your
program. In this case:
Why not:
-- assuming that there is an
-- instance ToJson Json where toJson = id
r - post site token user.addMedia $ Map.fromList
[(users, toJson users), (medias, toJson medias)]
Cheers!
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] ghc with -fPIC support
Hi All, I'm trying to install GHC with -fPIC support in Fedora. I've grabbed a source tarball since it seems no binary one has this. In Build.mk i've changed the quick build type to ifeq $(BuildFlavour) quick SRC_HC_OPTS= -H64m -O0 -fasm -fPIC GhcStage1HcOpts= -O -fasm -fPIC GhcStage2HcOpts= -O0 -fasm -fPIC GhcLibHcOpts = -O -fasm -fPIC SplitObjs = NO HADDOCK_DOCS = NO BUILD_DOCBOOK_HTML = NO BUILD_DOCBOOK_PS = NO BUILD_DOCBOOK_PDF = NO endif unfortunately, when compiling i still get the ld error *** *** ghc -fglasgow-exts --make -shared -oHs2lib.a /tmp/Hs2lib924498/Hs2lib.hs dllmain.o -static -fno-warn-deprecated-flags -O2 -package ghc -package Hs2lib -i/home/phyx/Documents/Haskell/Hs2lib -optl-Wl,-s -funfolding-use-threshold=16 -optc-O3 -optc-ffast-math Linking a.out ... *** /usr/bin/ld: /tmp/Hs2lib924498/Hs2lib.o: relocation R_X86_64_32 against `ghczmprim_GHCziUnit_Z0T_closure' can not be used when making a shared object; recompile with -fPIC *** /tmp/Hs2lib924498/Hs2lib.o: could not read symbols: Bad value I've also told cabal to build any packages with -fPIC and shared. Anyone have any ideas? Regards, Phyx ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANN: OpenCL 1.0.1.3 package
On Mon, Oct 3, 2011 at 9:04 AM, Jason Dagit dag...@gmail.com wrote: We could have a different version of the function for each return type, clGetDeviceInfo_FPConfig, clGetDeviceInfo_AddressBits, etc. It's a great naming convention but it has the property that someone searching the bindings or the bindings' haddocks for clGetDeviceInfo will find those functions. I meant to say, It's NOT a great naming convention ... Jason ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] ANN: OpenCL 1.0.1.3 package
On Mon, Oct 3, 2011 at 3:56 AM, Luis Cabellos cabel...@ifca.unican.es wrote: Hello, all. I want to show you the OpenCL package. I have done this using Jeff Heard OpenCLRaw package, but I create a new one due the lack of updates of the former. # Where to get it * Hackage page (http://hackage.haskell.org/package/OpenCL) * Repository (https://github.com/zhensydow/opencl) * Bugs (https://github.com/zhensydow/opencl/issues) * Examples (https://github.com/zhensydow/opencl/tree/master/examples). # Things: * I write it's high-level binding to OpenCL libraries, but only because I added more types to hide most of the alloc/free of the API, and hide the enums using c2hs enums. * The worst problem of the OpenCLRaw is the bad types it use, I learn to fix 32/64 bits issues with c2hs, and test it on linux machines. * Tested on Linux + NVidia only. * Jason Dagit is helping with Windows, OSX testing in own fork, also the call-conv fork in github has changes to work on Windows Your bindings are a higher quality than the the OpenCLRaw bindings and you're doing good technical work, but I stopped using your bindings for a couple reasons: * The main reason is that I'm not comfortable with the license you're using. The original code by Jeff Heard was BSD3 with an additional copyright notice. Your code is AGPL3. The GPL is known to cause problems with Haskell code due to cross module inlining. I don't know how the A in AGPL changes things. * Some of the exposed function names have been changed from the original name in the OpenCL specification. This is the same thing that was done with the OpenGL bindings and it is very confusing for people who come to the Haskell bindings from the official documentation. I realize that some of the API functions require some bit of name mangling, but I think the current way is not the right way. For example with this function: http://www.khronos.org/registry/cl/sdk/1.0/docs/man/xhtml/clGetDeviceInfo.html We could have a different version of the function for each return type, clGetDeviceInfo_FPConfig, clGetDeviceInfo_AddressBits, etc. It's a great naming convention but it has the property that someone searching the bindings or the bindings' haddocks for clGetDeviceInfo will find those functions. I think this is better than naming it clGetDeviceExtensions, which is not in the OpenCL specification. I'd still be willing to test the changes you have, I just don't want to contribute to your bindings due to the license. I currently thinking of starting my own bindings (Jeff's bindings contain too many small bugs and if I'm going to change most lines of code then I might as well start from scratch so that it can have a standard BSD3 license). Jason ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Installing hledger-web
On 10/2/11 10:12 AM, Arnaud Bailly wrote: No problem ! BTW, have you ever thought of coupling hledger with git for saving a ledger ? There is ongoing work to provide a native git interface. Yes, Clint Adams has begun adapting it to use the filestore (rcs abstraction layer) lib, and this is a goal of the 0.17 release. Additional help welcome! ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Trying to experiment with tangible values using GuiTV
Hello all,
I'm very new to Haskell, but still I'd like to play with tangible values,
which I find really intriguing. After getting some kind help in
haskell-beginners, I managed to compile and run a small example. Which is as
follows:
--
{-# LANGUAGE OverlappingInstances, UndecidableInstances
, IncoherentInstances, FlexibleContexts
, TypeSynonymInstances, FlexibleInstances
, MultiParamTypeClasses
#-}
-- For ghc 6.6 compatibility
{-# OPTIONS -fglasgow-exts -fallow-overlapping-instances
-fallow-incoherent-instances #-}
Some GuiTV examples. See also the examples in TV.
import Data.List (sort)
-- my addition
import Data.Monoid
-- end my addition
import Interface.TV.UI
import Control.Arrow.DeepArrow
import Data.FunArr
-- TypeCompose
import Data.Title
-- To pick up the FunArr instance for OFun.
import Interface.TV.OFun()
reverseT :: CTV (String - String)
reverseT = tv (oTitle reverse defaultOut) reverse
main = runUI reverseT
---
There were some warnings during compilation:
-
[1 of 1] Compiling Main ( myexample.hs, myexample.o )
myexample.hs:7:11:
Warning: -fallow-overlapping-instances is deprecated: use
-XOverlappingInstances or pragma {-# LANGUAGE OverlappingInstances #-}
instead
myexample.hs:7:11:
Warning: -fallow-incoherent-instances is deprecated: use
-XIncoherentInstances or pragma {-# LANGUAGE IncoherentInstances #-}
instead
Linking myexample ...
martin@martin-desktop:/media/
ext4logicaUnTera/cosas_linux/programs/haskell/TV/examples/src/miejemplo$
ghc --make myexample.hs
myexample.hs:7:11:
Warning: -fallow-overlapping-instances is deprecated: use
-XOverlappingInstances or pragma {-# LANGUAGE OverlappingInstances #-}
instead
myexample.hs:7:11:
Warning: -fallow-incoherent-instances is deprecated: use
-XIncoherentInstances or pragma {-# LANGUAGE IncoherentInstances #-}
instead
-
When I run it, I get a little window with the reverse title and two
textboxes. However, when I write something in the upper box, nothing
happens. I must say that when I use runIO instead of runUI in this little
program, the interactive program does work. Only the GUI version doesn't.
So, the person who helped me (Brandon Allbery) said it's probably because of
a GuiTV being a bit too old. He recommended me to ask about the issue here,
at haskell-cafe.
Any suggestions? Maybe I should use GtkTV istead of GuiTV? Thanks in
advance!
Martin O.B.
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Composing a list of Enumeratees into an Enumerator using ($=)
Hey guys, Right now I'm facing with a type problem that is really nasty, I want to compose a list of enumeratees using the ($=) operator to create a new enumerator. Whenever I'm trying to use the foldx function in conjunction with ($=) I get this error: * :t foldr ($=)* interactive:1:7: Occurs check: cannot construct the infinite type: b0 = Step ao0 m0 b0 Expected type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumeratee ao0 ao0 m0 b0 Actual type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumerator ao0 m0 b0 In the first argument of `foldr', namely `($=)' In the expression: foldr ($=) * :t Prelude.foldl ($=)* interactive:1:15: Occurs check: cannot construct the infinite type: b0 = Step ao0 m0 b0 Expected type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumerator ao0 m0 (Step ao0 m0 b0) Actual type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumerator ao0 m0 b0 In the first argument of `Prelude.foldl', namely `($=)' In the expression: Prelude.foldl ($=) interactive:1:15: Occurs check: cannot construct the infinite type: b0 = Step ao0 m0 b0 Expected type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumerator ao0 m0 (Step ao0 m0 b0) Actual type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumerator ao0 m0 b0 In the first argument of `Prelude.foldl', namely `($=)' In the expression: Prelude.foldl ($=) Obviously there is something I don't quite understand about the ($=) (=$) functions, how can one compose a list of enumeratees, is it even possible? Cheers. Roman.- ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Composing a list of Enumeratees into an Enumerator using ($=)
2011/10/4 Román González romanand...@gmail.com: Hey guys, Right now I'm facing with a type problem that is really nasty, I want to compose a list of enumeratees using the ($=) operator to create a new enumerator. Whenever I'm trying to use the foldx function in conjunction with ($=) I get this error: :t foldr ($=) interactive:1:7: Occurs check: cannot construct the infinite type: b0 = Step ao0 m0 b0 Expected type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumeratee ao0 ao0 m0 b0 Actual type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumerator ao0 m0 b0 In the first argument of `foldr', namely `($=)' In the expression: foldr ($=) :t Prelude.foldl ($=) interactive:1:15: Occurs check: cannot construct the infinite type: b0 = Step ao0 m0 b0 Expected type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumerator ao0 m0 (Step ao0 m0 b0) Actual type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumerator ao0 m0 b0 In the first argument of `Prelude.foldl', namely `($=)' In the expression: Prelude.foldl ($=) interactive:1:15: Occurs check: cannot construct the infinite type: b0 = Step ao0 m0 b0 Expected type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumerator ao0 m0 (Step ao0 m0 b0) Actual type: Enumerator ao0 m0 (Step ao0 m0 b0) - Enumeratee ao0 ao0 m0 b0 - Enumerator ao0 m0 b0 In the first argument of `Prelude.foldl', namely `($=)' In the expression: Prelude.foldl ($=) Obviously there is something I don't quite understand about the ($=) (=$) functions, how can one compose a list of enumeratees, is it even possible? Hi, what are you trying to actually do, ie. what kind of data are you trying to transform, what are the inputs and outputs of each enumeratee? are you trying to feed the output of the first enumeratee into the input of the second, and so on? or are you trying to run them all in parallel? Conrad. ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Installing hledger-web
Being a long time user of ledger though and a long time haskell fan, I would be more than happy to lend a hand but my free time is short. I will try to have a look to the sources and see if I can find opportunities to help. Regards, Arnaud On Mon, Oct 3, 2011 at 8:27 PM, Simon Michael si...@joyful.com wrote: On 10/2/11 10:12 AM, Arnaud Bailly wrote: No problem ! BTW, have you ever thought of coupling hledger with git for saving a ledger ? There is ongoing work to provide a native git interface. Yes, Clint Adams has begun adapting it to use the filestore (rcs abstraction layer) lib, and this is a goal of the 0.17 release. Additional help welcome! __**_ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/**mailman/listinfo/haskell-cafehttp://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] CAL or Frege as a Haskell replacement on JVM?
Hi Karel, I'm a (the? ;-) very keen user of CAL (ex user at the moment, as work and family doesn't leave me enough time for side projects). Pro: - Very solid and high quality, practically bug-free in my experience. - Performs some useful optimisations (self tail recursion as iteration, unboxing of primitives, strictness annotations/analysis) which can give you Java-level performance in some cases. - Good (IMHO) though verbose Java interop -- you can often turn Java types into CAL types rather than needing to wrap them in another layer. Con: - Haskell 98 type system (actually I found that a bit of a pro for learning fp) - Less syntactic sugar (do notation, equational definitions, pattern matching outside case statements) - Few users. There are occasional commits to forks of CAL at https://github.com/levans/Open-Quark and Luke Evans says that work on CAL continues at Indicee, but there's little visible activity. None of these points are in comparison to Frege, which I haven't used at all. Tom ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
