March 2010 - Beginners

Fwd: [Haskell-beginners] SMTPClient + hsemail-1.6 = errors
by Sergey Mironov 26 Mar '10

26 Mar '10

Thanks, Robert. But my real goal was to compile happstack. SMTPClient is just a dependency. I solved the problem with ghc-pkg unregister: ierton@vault ~ % ghc-pkg unregister hsemail-1.6 ierton@vault ~ % cabal install hsemail-1.3 ierton@vault ~ % caball install SMTPClient 2010/3/26 Robert Wills <wrwills(a)gmail.com>: > HaskellNet contains an smtp client as well. > > http://hackage.haskell.org/package/HaskellNet > > It worked last time I checked. > > main = do > con <- connectSMTP smtpServer > let msg = ([("From", "HaskellNet <" ++ sendFrom ++ > ">"),("Subject","Test")], BS.pack "\r\nhello from haskellnet") > sendMail sendFrom sendTo (BS.pack $ show $ showMessage "utf-8" msg) con > closeSMTP con > > Regards, > Rob > > On Thu, Mar 25, 2010 at 11:34 PM, Sergey Mironov <ierton(a)gmail.com> wrote: >> Hi. >> I was trying to install SMTPClient-1.0.1 >> >> Compiler reported me: >> >> ierton@vault SMTPClient-1.0.1 % cabal build >> Preprocessing library SMTPClient-1.0.1... >> Building SMTPClient-1.0.1... >> [1 of 2] Compiling Network.SMTP.ClientSession ( >> Network/SMTP/ClientSession.hs, dist/build/Network/SMTP/ClientSession.o >> ) >> >> Network/SMTP/ClientSession.hs:242:16: >> Not in scope: data constructor `Message' >> >> Network/SMTP/ClientSession.hs:406:15: >> Not in scope: data constructor `Message' >> >> I've made some digging: Seems that newest version of 'hsemail' (1.6) >> breaks backward compatibility (SMTPClient depends on it) >> >> My questions: >> 1) Where should i send bugreport to? >> There are no emails on http://hackage.haskell.org/package/SMTPClient >> so i don't know how to contact maintainers. >> >> 2) How to fallback hsemail? >> i tried: >> >> ierton@vault ~ % cabal install hsemail-1.3 # good, i suppose >> ierton@vault ~ % ghc-pkg hide hsemail-1.6 # Bad one >> ierton@vault ~ % caball install -v SMTPClient >> ... >> ... bla-bla ... >> ... >> selecting hsemail-1.6 (installed or hackage) and discarding hsemail-1.0, 1.1, >> 1.2, 1.3, 1.4 and 1.5 >> ... >> >> Looks like hiding a package is not a case for me. >> >> ps >> >> I am using ghc-6.12 >> >> -- >> Thanks, >> Sergey >> _______________________________________________ >> Beginners mailing list >> Beginners(a)haskell.org >> http://www.haskell.org/mailman/listinfo/beginners >> > _______________________________________________ > Beginners mailing list > Beginners(a)haskell.org > http://www.haskell.org/mailman/listinfo/beginners > -- С Уважением, Сергей. -- С Уважением, Сергей.

3 4

Cygwin error: D:\Haskell\bin/windres: can't open temporary file `\/cca02652.irc': No such file or directory
by remy willems 26 Mar '10

26 Mar '10

I am using cygwin. While in cygwin, when I try to make EHC, the essential haskell compiler, I get the following error: D:\Haskell\bin/windres: can't open temporary file `\/cca02652.irc': No such file or directory strip: 'build/libutil/setup.exe': No such file There are no strange permission in the folder I'm running the make. I'm the administrator and am running cygwin in "run as adminstrator" mode. Could someone help me? I'm really stuck with this. Much appreciated, Remy.

2 2

SMTPClient + hsemail-1.6 = errors
by Sergey Mironov 25 Mar '10

25 Mar '10

Hi. I was trying to install SMTPClient-1.0.1 Compiler reported me: ierton@vault SMTPClient-1.0.1 % cabal build Preprocessing library SMTPClient-1.0.1... Building SMTPClient-1.0.1... [1 of 2] Compiling Network.SMTP.ClientSession ( Network/SMTP/ClientSession.hs, dist/build/Network/SMTP/ClientSession.o ) Network/SMTP/ClientSession.hs:242:16: Not in scope: data constructor `Message' Network/SMTP/ClientSession.hs:406:15: Not in scope: data constructor `Message' I've made some digging: Seems that newest version of 'hsemail' (1.6) breaks backward compatibility (SMTPClient depends on it) My questions: 1) Where should i send bugreport to? There are no emails on http://hackage.haskell.org/package/SMTPClient so i don't know how to contact maintainers. 2) How to fallback hsemail? i tried: ierton@vault ~ % cabal install hsemail-1.3 # good, i suppose ierton@vault ~ % ghc-pkg hide hsemail-1.6 # Bad one ierton@vault ~ % caball install -v SMTPClient ... ... bla-bla ... ... selecting hsemail-1.6 (installed or hackage) and discarding hsemail-1.0, 1.1, 1.2, 1.3, 1.4 and 1.5 ... Looks like hiding a package is not a case for me. ps I am using ghc-6.12 -- Thanks, Sergey

2 1

Memory usage problem
by Sami Liedes 25 Mar '10

25 Mar '10

Hi! I wrote a simple program to parse Debian package list files (a simple text record format), sort the list of packages and output the list. I first wrote the program in C (which I speak fluently), but wanting to learn Haskell, have been trying to reimplement it in Haskell. The problem is that my 60-line Haskell program, which doesn't yet do everything the C program does, uses huge amounts of memory, 1.5 gigabytes when parsing a single file. The C implementation can parse and sort more than twice that data in less than 10 megabytes. I can't figure out where the memory goes or how to fix it, but my guess would be it's some lazy computation thunk. (See below after the code for description of the input and output formats.) With a package file of 28000 records (packages), the memory usage is roughly 1.5 gigabytes when compiled with ghc - under hugs it just bails out quickly after stack overflow - and I'd appreciate help in figuring out 1) Where the memory is actually spent? 2) How would one usually go about figuring this out? and 3) How to fix it? [4) I'd also love style hints and other ideas to make my code more idiomatic] I've tried to randomly add some $! operators to the code (that doesn't feel right :-), but so far to no avail. Here's the code: ------------------------------------------------------------ module Main where import List (sort) import Maybe (fromJust, mapMaybe) -- Parse one line of the format "Field: value". -- Ignore those that start with a space. readField :: String -> Maybe (String,String) readField line = if line == "" || head line == ' ' then Nothing else let (name,':':' ':val) = break (== ':') line in Just (name,val) data Package = Package { name, version :: !String } deriving (Show, Eq, Ord) -- Use sprintPackage for formatted output, show for unformatted sprintPackage :: Package -> String sprintPackage (Package name ver) = name ++ replicate (50 - length name) ' ' ++ ver -- Read one record worth of lines (separated by blank lines) getOneRecordLines [] = ([],[]) getOneRecordLines lines = (hd, if tl == [] then [] else tail tl) where (hd,tl) = break (== "") lines -- Read one record worth of lines and parse, returning (Field, value) -- tuples and the rest of the lines (less the just read record) readRecordFields :: [String] -> ([(String,String)], [String]) readRecordFields lines = (mapMaybe readField rl, rest) where (rl,rest) = getOneRecordLines lines -- Convert a list of (Field,value) tuples to a Package recordFieldsToPackage :: [(String,String)] -> Package recordFieldsToPackage fields = let names = [y | (x,y) <- fields, x=="Package"] vers = [y | (x,y) <- fields, x=="Version"] in case (names,vers) of ([], _) -> error "Package has no name." (a:b:rest, _) -> error ("Package has two names ("++a++","++b++").") ([n], []) -> error ("Package "++n++" has no version.") ([n], a:b:rest) -> error ("Package "++n++" has multiple versions.") ([n], [v]) -> Package n v -- Read one record, returning Just Package if it contained a valid -- package and version, Nothing otherwise. readRecord :: [String] -> (Maybe Package,[String]) readRecord lines = if fields == [] then (Nothing, rest) else (Just (recordFieldsToPackage fields), rest) where (fields,rest) = readRecordFields lines -- Converts the list (stream) of lines to a list of packages readRecords :: [String] -> [Package] readRecords [] = [] readRecords lines = let (rec,rest) = readRecord lines in case rec of Just pkg -> pkg : readRecords rest Nothing -> readRecords rest processFile = unlines . (map sprintPackage) . sort . readRecords . lines main :: IO () main = interact processFile ------------------------------------------------------------ The input (stdin) for the program is of the format: ------------------------------------------------------------ Package: somepackagename Version: someversionstring Other-Fields: ... Whatever: ... ... (lines that begin with space are just ignored) Package: pkg2 Version: otherversion ... ------------------------------------------------------------ That is, the file has records separated by blank lines. Only lines that begin with "Package: " or "Version: " are considered in a record, the others are just ignored. When finished, it outputs a formatted list of packages to stdout: ------------------------------------------------------------ a2ps 1:4.14-1 a2ps-perl-ja 1.45-5 a56 1.3-5 a7xpg 0.11.dfsg1-4 a7xpg-data 0.11.dfsg1-4 aa3d 1.0-8 aajm 0.4-3 aap 1.091-1 aap-doc 1.091-1 ... ------------------------------------------------------------ GHC profiling hints that the memory is spent in processFile and main, but no amount of adding $! to those functions seems to help: ------------------------------------------------------------ Thu Mar 18 22:37 2010 Time and Allocation Profiling Report (Final) aptlistsh +RTS -p -RTS total time = 3.18 secs (159 ticks @ 20 ms) total alloc = 1,873,922,000 bytes (excludes profiling overheads) COST CENTRE MODULE %time %alloc processFile Main 44.0 61.9 main Main 32.1 20.7 readField Main 11.9 9.0 recordFieldsToPackage Main 4.4 2.0 newPkg Main 3.1 0.8 sprintPackage Main 1.3 2.1 readRecordFields Main 1.3 1.3 CAF Main 1.3 0.4 getOneRecordLines Main 0.6 1.4 ------------------------------------------------------------ Sami

4 7

Call for papers (Deadline Extended): MULTICONF-10, USA, July 2010
by John Edward 24 Mar '10

24 Mar '10

It would be highly appreciated if you could share this announcement with your colleagues, students and individuals whose research is in computer science, computer engineering, information science and related areas. Call for papers (Deadline Extended): MULTICONF-10, USA, July 2010 The 2010 multi-conference (MULTICONF-10) (website: http://www.PromoteResearch.org ) will be held during July 12-14, 2010 in Orlando, Florida, USA. The primary goal of MULTICONF is to promote research and developmental activities in computer science, information technology, control engineering, and related fields. Another goal is to promote the dissemination of research to a multidisciplinary audience and to facilitate communication among researchers, developers, practitioners in different fields. The following conferences are planned to be organized as part of MULTICONF-10. • International Conference on Artificial Intelligence and Pattern Recognition (AIPR-10) • International Conference on Automation, Robotics and Control Systems (ARCS-10) • International Conference on Bioinformatics, Computational Biology, Genomics and Chemoinformatics (BCBGC-10) • International Conference on Computer Communications and Networks (CCN-10) • International Conference on Enterprise Information Systems and Web Technologies (EISWT-10) • International Conference on High Performance Computing Systems (HPCS-10) • International Conference on Information Security and Privacy (ISP-10) • International Conference on Image and Video Processing and Computer Vision (IVPCV-10) • International Conference on Software Engineering Theory and Practice (SETP-10) • International Conference on Theoretical and Mathematical Foundations of Computer Science (TMFCS-10) MULTICONF-10 will be held at Imperial Swan Hotel and Suites. It is a full-service resort that puts you in the middle of the fun! Located 1/2 block south of the famed International Drive, the hotel is just minutes from great entertainment like Walt Disney World® Resort, Universal Studios and Sea World Orlando. Guests can enjoy free scheduled transportation to these theme parks, as well as spacious accommodations, outdoor pools and on-site dining — all situated on 10 tropically landscaped acres. Here, guests can experience a full-service resort with discount hotel pricing in Orlando. Please see the website http://www.PromoteResearch.org for more details. Sincerely John Edward

1 0

web
by Dupont Corentin 24 Mar '10

24 Mar '10

Hello, what would you use to grab data displayed on a web site (several pages hyper-linked)? And then to compute and display statistics over that data? Thanks, Corentin

2 1

Applicative functors and trees
by Darryn Reid 23 Mar '10

23 Mar '10

Hi, I'm playing around with Functors in Haskell with the aim of understanding more structured functors, particularly applicative functors and traversable and foldable. I'm using the standard binary tree type to play around with this, as follows: data Tree a = Empty | Leaf a | Node (Tree a) a (Tree a) I think I understand Functor fine: instance Functor Tree where fmap f Empty = Empty fmap f (Leaf x) = Leaf (f x) fmap f (Node l x r) = Branch (fmap f l) (f x) (fmap f r) (with (<$>) being a synonym for fmap). This satisfies the laws for Functor (I'll omit details). Likewise, I think I'm fine with Foldable, for Tree a over any monoid a: instance Foldable Tree foldMap f Empty = mempty foldMap f (Leaf x) = f x foldMap f (Node l x r) = foldMap f l `mappend` f x `mappend` foldMap f r Where I'm stuck is with Applicative: class (Functor f) => Applicative f where pure :: a -> f a (<*>) :: f (a -> b) -> f a -> f b I understand that pure lifts a value into the applicative functor, so this one is easy: pure = Leaf but I cannot see what <*> should look like. I know that <*> is essentially application ($) lifted into the functor, so if the functor were also a monad then <*> would be liftM2 ($), I suppose. But I want a "direct" implementation, without monads floating about, but I cannot see what to do. I think then I would be able to define an instance for traversable as something like this: instance Traversable Tree traverse f Empty = pure Empty traverse f (Leaf x) = Leaf `fmap` f x traverse f (Node l x r) = Node `fmap` traverse f (l <*> f x <*> traverse f r) (I think that <*> must be associative). Can anyone advise what <*> should look like, please? Thanks in advance for your help. Darryn.

3 2

How to print the name of a variable
by John Velman 20 Mar '10

20 Mar '10

I have a number of variables (whose values are computed in my program) of the same type, and in ghci I want to be able to call a function to print something computed from the value of one or more of these to a file. I want the file name to reflect the name of the variable. This is probably not clear, but conceptually the function to do the output would look something like this: someFunction :: MyTypeVariable -> String writeFileFor :: MyTypeVariable -> IO () writeFileFor v = do nameOfV <- someFunction v outh <- openFile nameOfV ++ ".txt" WriteMode outputstring = calculationsFrom v hPutStr outh outputstring hClose outh Is there any such "someFunction", or is there a way to build one? Thanks, John Velman

3 6

ghci parse error on input
by Rick Murphy 20 Mar '10

20 Mar '10

Hello All: Using GHCi, I am working from an existing example which surprisingly contains a parse error in an export statement. I have reproduced the parse error as follows: > module Export > ( > (^) > , (#) > ) > where > import Prelude hiding ((^)) GHCi reports: parse error on input (# I read 5.2 Export Lists in the Haskell Report, but the syntax above does not appear to contradict the Haskell Report. I have eliminated the following solutions: 1. Add (#) to the hiding list. 2. Swap the (#) and (^). 3. Load as non-literate file. Would anyone be able to provide advice on how to avoid this parse error? -- Rick blog: http://phaneron.rickmurphy.org

3 4

Recursion/dynamic programming optimization
by ivovnenko 20 Mar '10

20 Mar '10

Hi. I tried to solve one of the google code jam's problems using haskel http://code.google.com/codejam/contest/dashboard?c=90101#s=p2: w :: Eq a => [a] -> [a] -> Int w text str = w' (reverse text) (reverse str) w' :: Eq a => [a] -> [a] -> Int -- Init w' _ [] = 0 w' [] _ = 0 -- For one symbol string w' text [a] = if head text == a then w' (tail text) [a] +1 else w' (tail text) [a] -- General case w' text str = if head str == head text then w' (tail text) (tail str) + w' (tail text) str else w' (tail text) str and it worked great on small inputs, not on large: w "viwiwhiuwwwgxzseebneesgzxiekzeeyivrbbhzrlllccxczyurxfhhccghkzvxusfbcucyvncpzpvfvccxcvbqnqogobzoiorxrbikyzzbobpmmmmpmvrzyivpfmmuzxixhqpeebyekkxffeyexeiukuebgbrukzefeu rkxy z h x vkq fsyp iy tfyzbptygtsrkhbytvruztzbttnoxuquokipyyqoik ib xu gxchcnccuccyfcccccsczccskifcprgovryxuxoozqoozsvudyfkdpkkxyhdvfsrnvpnpzvrhqhbduidkdpdysinyxnuxnxipddevqhgeufxeepyyezbekyviieeeryiexynevbzfeezryizvepynefffy y v zuzrbpzn sz y vzirkxjniuqvjjbjgubnjkujhjpiyanaabrqagzqqvspqpkvqyuhrgupbaiaasxybiygmmfimyvykkmrzum" "welcome to code jam" Can you please advice how to optimize this solution? Thanks. Regards, Ivan Vovnenko

3 2