March 2008 - Haskell-Cafe

No newlines in the whitespace for Parsec lexeme parsers
by Paul Keir 27 Mar '08

27 Mar '08

Hi, I'm looking to parse a Fortran dialect using Parsec, and was hoping to use the ParsecToken module. Fortran though is unlike the Parsec examples, and prohibits default line continuation (requiring instead an explicit ampersand token &). The lexical parsers of the ParsecToken module skip whitespace after each symbol parsed (lexeme parsers); including the newline. I was thinking of making a minor change to Parsec: In Token.hs, lexeme, whiteSpace, and simpleSpace are defined. lexeme uses whiteSpace which uses simpleSpace. simpleSpace is defined: simpleSpace = skipMany1 (satisfy isSpace) I could replace this locally with: simpleSpace = skipMany1 (satisfy isSpacePlusAmpersandMinusNewline) This approach may have other problems though. Has anyone experience of using Parsec to parse such a language? For example assembly or a preprocessor? Thanks in advance, Paul

4 4

FW: containers-0.1.0.1
by john lask 27 Mar '08

27 Mar '08

who is maintainer of containers package ... ---------------------------------------- > From: jvlask(a)hotmail.com > To: libraries(a)haskell.org > Subject: containers-0.1.0.1 > Date: Wed, 12 Mar 2008 23:48:33 +0000 > > > > containers-0.1.0.1 will not build with ghc-6.6, the lines > #include "Typeable.h" INSTANCE_TYPEABLE1(Set,setTc,"Set") > > cause a parsing error, due no doubt to Typeable.h being different to 6.8 version > _________________________________________________________________ > What are you waiting for? Join Lavalife FREE > _________________________________________________________________ Are you paid what you're worth? Find out: SEEK Salary Centre http://a.ninemsn.com.au/b.aspx?URL=http%3A%2F%2Fninemsn%2Eseek%2Ecom%2Eau%2…

1 0

Unboxed arrays
by Andrew Coppin 26 Mar '08

26 Mar '08

Somebody asked me, so now I'm asking you... In Haskell, you can make "unboxed" arrays of certain value types. These are typically more efficient in space, and probably time too, and also make the array strict in its values. However, you can only do this magic trick for certain types - not for *all* types. Why is that? Is it because nobody has done anything about it yet? Is it because it's thought to be "not necessary" in some way? Is there some theoretical problem? Has somebody got a better idea? I did think it was along the lines of "oh, well, if you want to unbox a type of your own, you just need to write your own instance". The thing that makes me suspicious of this logic is the absense of an instance for tuples. Surely this would be trivial to write, and yet it's not present. If we had instances for a couple of sizes of tuples, it would surely be quite easy to write your own custom instances that just sit on top of this and tuple/untuple your custom values... Any insights here?

7 13

Type synonyms
by Hugo Pacheco 26 Mar '08

26 Mar '08

Hi guys, There is something I think not to fully understand: what are the differences between these two type synonyms? type FInt x = Either One x type FInt = Either One Their kind is the same, so do they differ or are exactly the same type? Thanks, hugo

2 1

[GSoC] student applications deadline approaching
by Malcolm Wallace 26 Mar '08

26 Mar '08

Reply-To: haskell-cafe(a)haskell.org Just to remind students who are interested in Google Summer of Code. Student applications are now open, and the final deadline for submitting your proposals is 2400 UTC, 31st March. http://code.google.com/soc/2008/ Regards, Malcolm

1 0

Haskell in Web Browser: draft tutorial
by Dimitry Golubovsky 26 Mar '08

26 Mar '08

Hi, I have written a draft of the tutorial on Haskell programming for client side of Web applications (related to Yhc/Javascript backend): http://www.haskell.org/haskellwiki/Haskell_in_web_browser Any feedback, suggestions, and additions (like your own code examples) are welcome. Thanks. PS This part of the Tutorial does not cover XMLHTTP programming; this is yet to be written, but there is Haddock documentation under http://www.golubovsky.org:5984/_utils/yhcws/index.html Yhc web service online (where you may experiment with your code): http://www.golubovsky.org:5984/static/yhcws/MainGUI.html Users guide: http://www.haskell.org/haskellwiki/Yhc_web_service -- Dimitry Golubovsky Anywhere on the Web

1 0

Gtk2hs on GHC6.8.2?
by Ryan Ingram 25 Mar '08

25 Mar '08

The WinXP binary release of Gtk2hs won't install for me; it seems to expect GHC6.8.1 and refuses to install on 6.8.2. Is there any significant difference in the compilers that would cause it not to work on 6.8.2? Is there a way I can trick it into installing? I really don't want to have to build from source. I've had nothing but trouble trying to get large projects to build within cygwin/mingw/msys. -- ryan

2 2

Parsec (Zero or One of)
by Paul Keir 25 Mar '08

25 Mar '08

Hi, I'm having some difficulty using the Parsec library, perhaps you could help. I've reduced my problem as shown below. I would like the 'only_prod' parser to require the reserved string "only", _optionally_ followed by an identifier. As part of 'mytest', this should then be followed by the reserved string "end". mytest = do { only_prod; end } only_prod = do { reserved "only"; try identifier } end = reserved "end" i.e. I'd like both > parseTest mytest "only end" and >parseTest mytest "only green end" to parse successfully. As it stands, only the second is successful. The first fails with: parse error at (line 1, column 9): unexpected end of input expecting letter or digit or "end" Does anyone have an idea how I could repair the situation? Thanks in advance, Paul

4 7

True parallelism missing :-(
by Dusan Kolar 25 Mar '08

25 Mar '08

Dear all, I've thought the following three (dummy) programs would run some of their parts in parallel (on dual core) if compiled with option threaded (smp). The truth is that only the first one exploits multicore CPU. Why? Moreover, using RTS option -sstderr makes runtime not to evaluate in parallel even for the first program. Why? Thanks for tips Dusan My arch: Linux pcx 2.6.24-ARCH #1 SMP PREEMPT Sun Feb 10 15:44:59 CET 2008 x86_64 Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz GenuineIntel GNU/Linux My ghc: The Glorious Glasgow Haskell Compilation System, version 6.8.2 /64bit, binary distro for FC/ ---------------------------------------------------- Prog 1: module Main() where import Control.Parallel import Control.Parallel.Strategies fibs :: Integer -> Integer fibs n | n > 1 = fibs (n-1) + fibs (n-2) | n == 1 = 1 | True = 0 fib n = if n<0 then error "Negative input to fib!" else f1+f2 where [f1,f2] = parMap rnf fibs [(n-1),(n-2)] main = do putStrLn "Starting..." putStrLn $ "Fib 43: " ++ show (fib 43) putStrLn "Done!" ---------------------------------------------------- Prog 2: module Main() where import Control.Concurrent import Control.Concurrent.MVar fibs :: Integer -> Integer fibs n | n > 1 = fibs (n-1) + fibs (n-2) | n == 1 = 1 | True = 0 fib n = if n<0 then error "Negative input to fib!" else do v1 <- newEmptyMVar v2 <- newEmptyMVar h1 <- forkIO $ putMVar v1 $ fibs (n-1) h2 <- forkIO $ putMVar v2 $ fibs (n-2) f1 <- takeMVar v1 f2 <- takeMVar v2 killThread h1 killThread h2 return (f1+f2) main = do putStrLn "Starting..." f <- fib 43 putStrLn $ "Fib 43: " ++ show f putStrLn "Done!" ---------------------------------------------------- Prog 3: module Main() where import Control.Concurrent import Control.Concurrent.MVar fibs :: Integer -> Integer fibs n | n > 1 = fibs (n-1) + fibs (n-2) | n == 1 = 1 | True = 0 fib n = if n<0 then error "Negative input to fib!" else do v1 <- newEmptyMVar v2 <- newEmptyMVar h1 <- forkOS $ putMVar v1 $ fibs (n-1) h2 <- forkOS $ putMVar v2 $ fibs (n-2) f1 <- takeMVar v1 f2 <- takeMVar v2 killThread h1 killThread h2 return (f1+f2) main = do putStrLn "Starting..." f <- fib 43 putStrLn $ "Fib 43: " ++ show f putStrLn "Done!"

4 5

ANNOUNCE: Haskell Server Pages v 0.4
by Niklas Broberg 25 Mar '08

25 Mar '08

Greetings fellow Haskelleers, I am very pleased to announce a new chapter in the Haskell Server Pages saga. After a two-year long hiatus, while we in the team have been busy with Other Stuff, we have resumed work on Project HSP, and this release marks the first milestone. ================= === Project HSP === ================= Haskell Server Pages (HSP) is a programming model for writing dynamic web pages in Haskell, both server-side and client-side. One of its core features is the use of literal XML syntax mixed with ordinary Haskell code. * Project HSP home (with bug/feature tracker): http://code.google.com/p/hsp * Project HSP mailing list: haskell-server-pages(a)googlegroups.com * Project HSP darcs repos: http://code.haskell.org/HSP For flavor, here is a simple Hello World application written in HSP: ----- module HelloWorld where import HSP import HSP.HJScript import HJScript import HJScript.DOM -- A simple HTML page holding a button with onClick action. page :: HSP XML page = <html> <head> <title>Hello World</title> </head> <body> <% <button>click</button> `onClick` (window # alert (string "Hello World !")) %> </body> </html> ----- Note that literal XML elements are first class expressions in HSP, unlike languages like PHP. Also note that using literal XML syntax is by no means required, if you don't feel like writing the XML manually (I sure don't!). There are combinators for building XHTML pages (although they are currently missing from version 0.4, will be added shortly, but release early and all that...). The most interesting new feature in version 0.4 is that literal XML syntax (or the equivalent combinators) can now be used to create XML elements both server-side and client-side, with the same code. Here is another short example script showing this feature: ------------------ module AddParagraph where import HJScript hiding (test) import HJScript.DOM import HSP import HSP.HJScript makeP x = <p><b><% x %></b></p> page :: HSP XML page = <html> <head><title>Add paragraph</title></head> <% do (r,b) <- ref <body><% makeP "Paragraph created Server-side!" %></body> let fun = do let x = document # getElementById (string r) x <: makeP "Paragraph created Client-side!" return () b <: (<button>Click Me!</button> `onClick` fun) %> </html> -------------------- Project HSP v 0.4 consists of a set of packages with related functionality: == package hsx == Haskell Source with XML (HSX, hsx) is a package that contains everything pertaining to literal XML syntax. In particular it contains a) the trhsx preprocessor that translates hsx source files into vanilla Haskell, and b) modules defining the functions that are injected by trhsx. It also nominally contains generic combinators for creating values of the same types as the literal XML syntax, though these modules are not present in 0.4. darcs get --partial http://code.haskell.org/HSP/hsx http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hsx-0.4 == package hsp == The core HSP package defines the datatypes and functions for writing server-side dynamic web pages. Also defines how to use the HJScript functionality in HSP pages, to allow for client-side dynamics as well. darcs get --partial http://code.haskell.org/HSP/hsp http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hsp-0.4 == package HJavaScript == This package defines an abstract syntax and (not-so-)pretty printer for a large subset of JavaScript, as Language.HJavaScript. However, a significant difference from JavaScript is that HJavaScript is typed, even on the abstract syntax level using GADTs. The subset of JavaScript that is supported is those parts that lend themself to typing (i.e. no prototyping of classes). darcs get --partial http://code.haskell.org/HSP/hjavascript http://hackage.haskell.org/cgi-bin/hackage-scripts/package/HJavaScript-0.4 == package HJScript == HJScript is a DSL built on top of HJavaScript, for writing client-side dynamic web pages. The programming model is fairly low-level, resembling the actual JavaScript code quite a lot, but should be easy to extend with higher-level functionality. Notable is that HJScript supports the use of literal XML syntax for creating DOM ElementNodes. Also notable is that HJScript supports Ajax functionality. HJScript and HJavaScript can be used independently of HSP, for creating JavaScript code. darcs get --partial http://code.haskell.org/HSP/hjscript http://hackage.haskell.org/cgi-bin/hackage-scripts/package/HJScript-0.4 == package hsp-cgi == Run HSP pages as CGI scripts. darcs get --partial http://code.haskell.org/HSP/hsp-cgi http://hackage.haskell.org/cgi-bin/hackage-scripts/package/hsp-cgi-0.4 (Note: Earlier versions of HSP have come with a server application, hspr, providing the runtime environment for HSP. As of 0.4, this server has been discontinued and should be considered deprecated.) ===================== Notable items on the TODO list for Project HSP: - More functionality, allowing for higher-level programming models. - Writing larger applications, in order to stress test, and serve as inspiration for higher-level functionality. - Integrate with HAppS, to allow stateful computation (among other things). - A home on the web for HSP, with a web page written in HSP. We welcome both users (of course) and contributors to the project. There is much work to be done, and we welcome all help we can get. If you think this is interesting stuff, drop by our mailing list (haskell-server-pages(a)googlegroups.com) Like to do web programming in Haskell? Want to do a GSoC project? We in the project would be happy to see applications for HSP projects, and the list above is a good starting place. See for instance (http://hackage.haskell.org/trac/summer-of-code/ticket/1538) and (http://hackage.haskell.org/trac/summer-of-code/ticket/1539) Cheers, /Niklas on behalf of the Project HSP team

1 0