November 2009 - Haskell-Cafe

Sometimes pinned memory?
by Gregory Crosswhite 13 Nov '09

13 Nov '09

Hey everyone! Do you have any suggestions for how I might allocate an aligned block of memory that I can pin while making foreign calls, but leave unpinned the rest of the time to potentially improve allocation and garbage collector performance? Or is this even a good idea? Thanks, Greg

4 5

Resource compilation in GHC
by Konstantin Vladimirov 13 Nov '09

13 Nov '09

Hello. I'm writing an wxHaskell application. Everything is ok, but now I need a separate folder for icons, bitmaps, and so on, from where they are loaded at runtime. How can I compile resources, and link them into my executable to provide for users single .exe file with resource section inside it? -- With best regards, Konstantin

4 3

C structures
by Vasiliy G. Stavenko 13 Nov '09

13 Nov '09

Hello everyone. What about passing complex c-types (structures) to c-functions. More detailed: I have an application in production which was written in Delphi. IT has ability to create pluggable modules to it. Interface realized by sending Win32Api messages to application. function in haskell Win32 Graphics.Win32.sendMessage :: HWND -> WyidowMessage-> WPARAM -> LPARAM -> IO RESULT Application wants to get different data in WPARAM and LPARAM. Such as window descriptor (HWND), string pointers and datastructer pointers. The latter i don't know how to create. I need anyones experience.

2 2

The weirdest error I've ever seen...
by Joe Fredette 13 Nov '09

13 Nov '09

Hiya Haskellers, So there I was, punching away at the keys, working on the Haskell Weekly News tools when the solution to one of my problems fell on me like a ton of lambdas. The solution and problem it solved are immaterial, but suffice to say it involved the combination of associated types and monad transformers, as well as some fancy deriving to end up with this code: ##### type Context = ReaderT Email type Match t = StateT t IO type ContextMatch t a = Context (Match t) a newtype FilterState t => Filter t a = Filter (ContextMatch t a) deriving (Functor, Monad, MonadReader Email, MonadState Bool, MonadIO) class FilterState t where data FState t deliver :: FState t -> IO () ##### Again, the fine details are unimportant, but the punchline is `Filter` is a Monad which houses not only results, but also an internal state which will be used in the delivery of emails in some yet-to-be- determined way. Naturally, I want to use `deriving` to turn this puppy into a monad over it's second argument. In fact, the whole thing kind- checks alright, but presents me with this, the titular 'weirdest error I've ever seen...' ##### [1 of 3] Compiling Network.HackMail.Email.ParseEmail ( Network/ HackMail/Email/ParseEmail.hs, interpreted ) [2 of 3] Compiling Network.HackMail.Email.Email ( Network/HackMail/ Email/Email.hs, interpreted ) [3 of 3] Compiling Network.HackMail.Filter.Filter ( Network/HackMail/ Filter/Filter.hs, interpreted ) *** Exception: No match in record selector Var.tcTyVarDetails ##### Now, there are three tickets open on the GHC trac, found for me by the ever-helpful `copumpkin` on #haskell -- because I didn't think to look -- they are numbers 3621, 3422 and 2714. But none of them are sufficiently close to my case for them to make sense to me, nor are the solutions presented suitable for entry into my feeble noggin. (Thats just a purty way of saying I'm not smart enough to understand what any of it means...) So I beseech my fellow Haskellers[1], What the heck did I do to anger the Var.tcTyVarDetail gods? My guess (given what I can glean from the Trac entries) is that the `deriving ... MonadState ...` needs changing in some specific-yet- cryptic way, but I've only got my gut to go on... For the Record, and in the event it matters... [jfredett@Erdos]$ ghc --version The Glorious Glasgow Haskell Compilation System, version 6.10.4 [jfredett@Erdos]$ uname -a Linux Erdos 2.6.31-ARCH #1 SMP PREEMPT Fri Oct 23 11:12:58 CEST 2009 i686 Intel(R) Celeron(R) CPU 3.06GHz GenuineIntel GNU/Linux Thanks in advance for any help offered. /Joe [1] Bet you've never been beseeched before...

3 5

flow2dot
by Anakreon Mendis 12 Nov '09

12 Nov '09

I've installed the flow2dot utility. It fails to produce a dot file from the sample provided by it's author. The output of the program is: .cabal/bin/flow2dot sample.flow flow2dot: Input: order a b c d a -> b: let's play "catch a ball"! b -> c: i'll pass it along c: what to do next? c -> a: lets throw it back again a: "not fair!" a -> c: find someone else! c -> d: ok, here you go: catch! d: zZzZzZzZzzz.... Error: "sample.flow" (line 1, column 1): unexpected "o" expecting "->", ":" or end of input

2 1

Why can I run (runghc) some Haskell scripts but I cannot seem to compile them?
by Casey Hawthorne 12 Nov '09

12 Nov '09

Why can I run (runghc) some Haskell scripts but I cannot seem to compile them? e.g. http://www.haskell.org/all_about_monads/examples/example25.hs I've changed the import listing to the following: import IO import System import Monad import Data.Maybe import Data.List import Data.Char (toLower) import Control.Monad.State import Control.Monad.Writer The compiler errors are like the following: N-Queens.o:fake:(.text+0x2fe): undefined reference to `mtlzm1zi1zi0zi2_Controlzi MonadziStateziLazzy_zdf9_closure' N-Queens.o:fake:(.text+0x422): undefined reference to `mtlzm1zi1zi0zi2_Controlzi MonadziStateziLazzy_zdf7_closure' N-Queens.o:fake:(.text+0x66f6): undefined reference to `mtlzm1zi1zi0zi2_Controlz iMonadziWriterziLazzy_zdf1_closure' -- Regards, Casey

2 2

Why can I run (runghc) some Haskell scripts but I cannot seem to compile them?
by Casey Hawthorne 12 Nov '09

12 Nov '09

Why can I run (runghc) some Haskell scripts but I cannot seem to compile them? e.g. http://www.haskell.org/all_about_monads/examples/example25.hs I've changed the import listing to the following: import IO import System import Monad import Data.Maybe import Data.List import Data.Char (toLower) import Control.Monad.State import Control.Monad.Writer -- Regards, Casey

1 0

Linker?
by Gregory Crosswhite 12 Nov '09

12 Nov '09

Does anyone know here how GHC links in object files from other languages? I am getting a strange issue where it seems to be getting the calling convention on Fortran calls wrong. Specifically, on one computer (Gentoo Linux) I have with gcc and gfortran v4.4 and ghc compiled using gcc v4.3, I have no problems at all. On another computer (a Mac) I have with gcc 4.0 and gfortran 4.4 and a binary of ghc installed (as downloaded from haskell.org) I run into problems. Specifically, I noticed that the first parameter seems to get eaten when I declared a fortran routine as a function but not when I declare it as a subroutine. So I went ahead and converted the fortran routines into subroutines, but then I get another crash elsewhere in the code where a module calls a LAPACK routine for reasons unknown; I do not get this crash when running the same code path but called using a Python wrapper that uses gcc + gfortran + f2py rather than from my Haskell wrapper. I thought that this might be a problem with mixing gcc 4.0 and gfortran 4.4 (and ultimately I think this might still be the case), but when I tried writing small test cases using *only* gcc and gfortran and not ghc, everything worked just fine. Any thoughts? - Greg

1 0

Re: [Haskell-cafe] looking for a good algorithm
by Casey Hawthorne 12 Nov '09

12 Nov '09

To: Casey Hawthorne <caseyh(a)istar.ca> Subject: Re: [Haskell-cafe] looking for a good algorithm From: Casey Hawthorne <caseyh(a)istar.ca> Date: Thu, 12 Nov 2009 11:14:02 -0800 On third thought, convert the table to a 2D array of bits (or a 1D array of bits mapped to a 2D coordinate system). The bit is true for either an Int or Double. If space is a concern and one can tolerate a low rate of false positives, than a 2D Bloom Filter might be appropriate, or a 1D Bloom Filter with mapping to 2D coordinates. http://en.wikipedia.org/wiki/Bloom_filter Then on to simmulated annealing, possibly. Note: Donald Knuth's new fascicle is out: Volume 4 Fascicle 1, Bitwise Tricks & Techniques; Binary Decision Diagrams (2009), xiii+261pp. ISBN 0-321-58050-8 On Wed, 11 Nov 2009 15:32:35 -0800, you wrote: >So, as I understand it, you have a very large sparse table, thousands >of rows and hundreds of columns, of which each cell within a column of >type String, Int, or Double can contain one of those types or nothing. > >Then you to want to shuffle the rows to maximize the number of columns >whose first 100 rows have at least one number (Int or Double), given a >list of preferred column names since there is no guarantee that every >number column will have at least one number in its first 100 rows >after shuffling. > > >I'm wondering about hashing on the rows and hashing on the columns, >then the column hash has the number of Int's or Double's (don't need >the String's) in that column and the rows they are in. > >The row hash would have the number of Int's and Double's in that row >and what column's they are in. > >Then; > >Then scan the row hash and sort into descending order, and by tagging >those rows, not by actually moving them. > >Then I think your ready for simmulated annealing. -- Regards, Casey

1 0

Cabal upload issue
by Jeremy O'Donoghue 12 Nov '09

12 Nov '09

Hi all, I'm in the process of trying update the revisions of wx (part of wxHaskell) on hackage. I'm getting an error I find slightly surprising: "400 Error in upload The dependency 'build-depends: base' does not specify an upper bound on the version number. Each major release of the 'base' package changes the API in various ways and most packages will need some changes to compile with it. The recommended practise is to specify an upper bound on the version of the 'base' package. This ensures your package will continue to build when a new major version of the 'base' package is released. If you are not sure what upper bound to use then use the next major version. For example if you have tested your package with 'base' version 2 and 3 then use 'build-depends: base >= 2 && < 4'." In this case, we have the following in wx.cabal Library if flag(splitBase) build-depends: base >= 3, wxcore >= 0.12.1.1, stm else build-depends: base < 3, wxcore >= 0.12.1.1, stm Is this a bug, or am I doing something stupid? I should point out that this builds perfectly on my machine. Regards Jeremy

4 3