March 2010 - Haskell-Cafe

Re: [Haskell-cafe] Books for "advanced" Haskell
by Günther Schmidt 06 Mar '10

06 Mar '10

Hi John, please, do try. Günther

1 0

How to obtain unique integers
by Robert Rothenberg 06 Mar '10

06 Mar '10

I'm implementing a variant of Prolog in Haskell (yes, I know others already exist...) and am looking to solve the following issues: (1) I need to translate anonymous variables "_" into unique variable names. Data.Unique does not give me a printable identifier. Is there another package that does this? (2) Is there a cabal package similar to the Atom.hs module form ginsu, that associates strings with unique hashes? The alternative of course is to maintain state variables and pass them along to display and parsing functions, but that makes it difficult to make the terms instances of Read/Show without reference to a global variable anyhow. Thanks, Rob

3 3

Re: Gut Build System
by Achim Schneider 05 Mar '10

05 Mar '10

Matthias Kilian <kili(a)outback.escape.de> wrote: > On Fri, Mar 05, 2010 at 09:26:08PM +0000, Simon Marlow wrote: > > >Switching GHC to a completely new build system written completely > > >in Haskell would be the most stupid idea ever. (You know why) > > > > You're referring to bootstrapping, I presume? > > Yes. > As it stands, GHC can't be bootstrapped without a GHC, anyway. I'd love to see GHC bootstrapping from C, e.g. via hugs, though. (A man is allowed to dream) > > I did think about mentioning that. Of course any Haskell build > > system would have to support cross-compilation or some other > > bootstrapping method, e.g. generation of a build script. > > That's a noble goal. But it has to be kept in mind. > This may be fairly easy, if the generated script does not have to come with major build-system capabilities: Leaving out most of dependency analysis and support -- at the utmost -- restarting a failed build from the rule whose source file has last recently changed, otherwise compiling in a linear order deduced at script generation time should reduce the problem to serializing the actions generated for a clean build to a shell script. I'm cross-posting this to the cafe and suggest that most if not all of this discussion be done there as most of it doesn't specifically relate to GHC. And there might be volunteers amidst the crowd. That said, I'd very much like to see a make-replacement done in Haskell, there's not only place for a properly-scriptable make, but also sufficient room for general innovation: Omake[1] would be even better if it wasn't scripted in some ad-hoc, impure, incomplete ocaml dialect but in a decently pure functional language; IBB[2] sound like a feature we'd definitely want to have (thanks for the link, Thomas) I'd also like to point to Nix[3] and Cabal[4] for ideas, and I bet most of the dependency analysis could be ripped from the latter. I think we can do vastly better than the whole SCons/CMake/Jam crowd. [1] http://omake.metaprl.org/index.html [2]http://chadaustin.me/2010/03/your-version-control-and-build-systems-dont-scale-introducing-ibb/ [3] http://nixos.org/nix/ [4] You know where. -- (c) this sig last receiving data processing entity. Inspect headers for copyright history. All rights reserved. Copying, hiring, renting, performance and/or quoting of this signature prohibited.

3 3

Using Haskell's FFI to send ancillary data over Unix domain sockets
by ihope 05 Mar '10

05 Mar '10

Unix domain sockets are a type of socket created between two programs on a single Unix system. They are useful in part because over them you can send so-called ancillary data: file descriptors and credentials (i.e. a proof of who the process on the other end is). The thing is, Haskell doesn't have a nice way of sending ancillary data. Network.Socket does have these really opaque functions for sending and receiving ancillary data: sendAncillary :: Socket -> Int -> Int -> Int -> Ptr a -> Int -> IO () sendAncillary sock level ty flags datum len = do ... recvAncillary :: Socket -> Int -> Int -> IO (Int, Int, Ptr a, Int) recvAncillary sock flags len = do ... return (lev,ty,pD,len) Looking in the man page UNIX(7), which describes Unix domain sockets, some enlightening information is given. It says that ancillary data is sent and received using sendmsg(2) and recvmsg(2). Those two man pages say that sock and flags are arguments to sendmsg and recvmsg. UNIX(7) says that level is always SOL_SOCKET, and ty is either SCM_RIGHTS or SCM_CREDENTIALS depending on whether the ancillary data is file descriptors or credentials. In the former case, datum is "an integer array of the file descriptors"; in the latter case, it's the following struct: struct ucred { pid_t pid; /* process ID of the sending process */ uid_t uid; /* user ID of the sending process */ gid_t gid; /* group ID of the sending process */ }; As for len, this struct seems to be enlightening: struct cmsghdr { socklen_t cmsg_len; /* data byte count, including header */ int cmsg_level; /* originating protocol */ int cmsg_type; /* protocol-specific type */ /* followed by unsigned char cmsg_data[]; */ }; So, sock is a Socket, level is the constant SOL_SOCKET, ty is either SCM_RIGHTS or SCM_CREDENTIALS (strangely, Network.Socket contains the former but not the latter), flags is the flags passed to sendmsg or recvmsg (whatever those are--what are they?), datum is either a C integer array or that struct, and len is the length in bytes of all that. For me, this presents a few problems. I don't know where to get the SCM_CREDENTIALS constant, I have no idea what flags to use (does the Network module help with that?), I don't know how to get from a list of file descriptors or a tuple to a Ptr, and perhaps most importantly, I have no idea how to get the lengths of pid_t, uid_t, gid_t, and socklen_t. Can anyone offer assistance?

2 1

GPL answers from the SFLC (WAS: Re: ANN: hakyll-0.1)
by Kevin Jardine 05 Mar '10

05 Mar '10

I'm a Haskell newbie but long time open source developer and I've been following this thread with some interest. The GPL is not just a license - it is a form of social engineering and social contract. The idea if I use the GPL is that I am releasing free and open source software to the community. You are welcome to use it for any purpose but in exchange you must also agree to release any software you create that uses my software as free and open source. That is the difference between GPL and BSD type licenses. The GPL very deliberately creates an obligation. Yes, that can be inconvenient. It is meant to be inconvenient. Actually the GPL reminds me of a Haskell concept that I am struggling with right now - the monad. When I started writing Haskell code I was always trying to mix pure and IO code and I soon learned that once I used the IO monad I was stuck within it. The monad creates an inconvenient obligation and any IO code can only be used within other IO code. There are good reasons for monads (just as, in my view, there are good reasons for the GPL) but using them means that I need to make a lot of changes to the way I write software. Kevin

6 5

How to put data from a string to a tuple
by Pradeep Wickramanayake 05 Mar '10

05 Mar '10

As luke requested Here is the codes and some more brief explanation. Please help me getItemFile :: IO String getItemFile = do test <- readFile "input.txt" return test im taking a file data to a string it contains string like "hello,world,I,am,a,new,developer" so I need to put these each to a tuple. Because the content have Integers and Strings sortList2 :: String -> String sortList2 (x:xs) | x == ',' = "" | otherwise = [x] ++ sortList2 xs Im breaking word by word from this above function Now I need to send it to a tuple. Can someone help me how to do it. Is this possible. Using recursion im checking each word till "," occurs And taking that string and passing it to the tuple. Can someone please help me to do it. Please

2 1

New OpenGL package: efficient way to convert datatypes?
by Peter Verswyvelen 05 Mar '10

05 Mar '10

The newest package seems to require using GLdouble/GLfloat. What is the most efficient way to convert Double/Float to GLdouble/GLfloat? I'm currently using realToFrac. But essentially the operation should be a nop on my machine. I haven't looked at the core code yet (on Windows, last time I checked, the ghc-core util did not work, and without it, reading core is even harder) Thanks, Peter

7 13

Hackage vs. legacy support in Cabal files
by John D. Ramsdell 05 Mar '10

05 Mar '10

I'd like to support Hackage and GHC 6.8 with one source distribution. The enclosed Cabal file would do the trick if Hackage would simply evaluate the Cabal file using the default flag values. Instead, it chokes on a hack required to support GHC 6.8. The directory containing the autogen files has to be on the source path. if flag(old) Hs-Source-Dirs: src dist/build/autogen else Hs-Source-Dirs: src Is there a way to force Hackage to respect flag values? John

1 0

Re: A few ideas about FRP and arbitrary access in time
by Patai Gergely 05 Mar '10

05 Mar '10

> I'm not sure I want to follow WWRD all the way. I do want events, for > example mouse clicks (for which there doesn't seem to be any logical > behavior representation). As you note the pull-based approach does a > lot more work in those cases than seems necessary. Events are indispensable for practically all applications, even those that have continuous output, so leaving them out would be no option. I only did so for Elerea to be able to quickly get a working system running, so I can actually experiment with application-side design. But yeah, it can't be used for serious purposes without some support for events. > In any case, what I'm discussing in that report is mostly centered on > the continuous-time values, so I'm more concerned with what we call > behaviors. The question is how to model memory-full operations with > point-wise operations. I think it would help if you tried to solve more specific problems first and see what common structure you can factor out from the solutions. It's easy to get into a dead end if you go for the big picture right away. > Right. You'll agree that it would be nice to have a general approach? > How will that work? That's what I'd like to find out. And I do hope you'll manage to find a useful result. My intuition says, however, that looking for a sensible interface between events and behaviours is more fruitful than trying to unify these two notions. The reason is the difference in TCMs, as noted previously. > report. I'd like things to be unified, so maybe playing with the idea > of a total function to 'Maybe a' for events is the right direction. I > think this was already explored by some FRP incarnations, but I don't > recall what came out of it. Yes, that's the Yampa way. The problem is that in this case you'll need an awkward operator to refer to 'the time of last defined point'. > But isn't Lucid Synchrone essentially discrete-timed? Also, events > shouldn't be semantically constrained to multiples of some basic > clock, they are defined over continuous time. Of course, it's more like a possible model for operational semantics (I know you're interested in denotation, but inspiration can come from anywhere). I just thought that clock calculus might be an interesting topic to throw into the hat. Gergely -- http://www.fastmail.fm - Email service worth paying for. Try it for free

1 0

How to put data from a string to a tuple
by Pradeep Wickramanayake 05 Mar '10

05 Mar '10

Hi, Im self learner in Haskell. And im stuck in a small place which I tried searching in google but couldn't find Proper answer I have some values in a string, Im removing them one by one from the string (word by word) and I want to put them in a tuple. Because it contain Integers and Strings. Can someone help me __________ Information from ESET NOD32 Antivirus, version of virus signature database 4917 (20100305) __________ The message was checked by ESET NOD32 Antivirus. http://www.eset.com

3 2