- Haskell-Cafe - Haskell.org

Network.CGI strangeness
by Alexander Solla 21 Jul '11

21 Jul '11

Hi Everybody, I'm having some strange issues with Network.CGI's pathInfo action, or else some other strangeness with pattern matching. My "mainCGI" CGI action is: mainCGI :: CGI CGIResult mainCGI = pathInfo >>= outputSlotTHtml . packagesPage outputSlotTHtml is irrelevant to my problem. packagesPage generates a SlotT IO Html -- Slot semantics are essentially irrelevant to my problem. But packagesPage calls a function called pathCrossRef which is having some problems: packagesPage :: WebPath -> SlotT IO Html packagesPage wp = do infos <- getPackageInfos . pathCrossRef $ wp ... pathCrossRef :: String -> String pathCrossRef "/working/" = "/development/code/haskell/working/" pathCrossRef _ = "/development/code/haskell/packages/" The intention is that when the cgi program is called with "/working/" as an argument, the pathCrossRef should point at "/development/.../working/", and when called with "/packages/" (or anything else) as an argument, pathCrossRef should point to "/development/.../packages/". But it is not working! When I visit http://localhost/working/, I end up using the "packages" cross ref. I have even printed the cgi argument to the Apache log, and the argument is "/working/". If I get rid of the wildcard pattern, I get a "missing pattern" error when I visit http://localhost/working/. Can anybody help? Thanks, Alex

1 0

ANN: boomerang and web-routes-boomerang
by Jeremy Shaw 21 Jul '11

21 Jul '11

Hello, I am pleased to announce the release of two new libraries: boomerang and web-routes-boomerang. boomerang is a library for general purpose, invertible parsing and pretty printing. It provides combinators which allow you to specify a grammar once and automatically extract a parser and pretty-printer from it. This library was refactored, with permission, from the Zwaluw library created by Sjoerd Visscher and Martijn van Steenbergen. hackage: http://hackage.haskell.org/package/boomerang quick intro: http://hackage.haskell.org/packages/archive/boomerang/1.0.0/doc/html/Text-B… web-route-boomerang provides glue code for using boomerang with web-routes. web-routes is a framework-independent, type-safe, url routing library. hackage: http://hackage.haskell.org/package/web-routes-boomerang tutorial: http://happstack.com/docs/crashcourse/WebRoutes.html#web-routes-boomerang - jeremy p.s. boomerang is similar in purpose to the inveritble-syntax library: http://hackage.haskell.org/package/invertible-syntax The libraries take different approaches and are both worth considering.

2 3

Using SAX
by Larry Lee 21 Jul '11

21 Jul '11

I'm new to Haskell and am trying to use the SAX module to parse an XML file. I'm using the following code: module FPHParser where import Data.Text as Text import Data.Maybe import Text.XML.LibXML.SAX as SAX import System.IO f :: IO (Parser IO) f = do let g = do putStrLn ("g called.") return True let h msg = do putStrLn (Text.unpack msg) return False p <- SAX.newParserIO (Just (Text.pack "test.xml")) SAX.setCallback p SAX.reportError h SAX.setCallback p SAX.parsedBeginDocument g SAX.parseComplete p return p and am getting the following error when I parse a file with the following contents: <?xml version="1.0"?> <html/> [This is the error] "Extra content at the end of the document" This error seems to be coming from the LibSAX library, which is a C library. But I think that my error is in my code, as the file is valid XML. Any help would be greatly appreciated!

1 0

Make Show instance
by Александр 21 Jul '11

21 Jul '11

Hello, I have binary tree, with every leaf tuple - (k,v): data Tree k v = EmptyTree | Node (k, v) (Tree k v) (Tree k v) How can i make Show Instance for (Tree Int Int) ? I try: instance Show (Tree k v) where show EmptyTree = show "Empty" show (Node (Int, Int) left right) = .. But get error: Not in scope: data constructor `Int' I have a function: fillTree :: Int -> Tree Int Int -> Tree Int Int fillTree 1000000 tree = tree fillTree x tree = let a = treeInsert (x, x) EmptyTree in fillTree (x + 1) a If i execute it: No instance for (Show (Tree Int Int)) arising from a use of `print' How can i make Show instance for my Tree? Thank you.

4 5

Helsinki Haskellers?
by Dean Herington 21 Jul '11

21 Jul '11

Hi, Cafe, My wife, a biologist, will attend a conference outside Helsinki in mid-September. I've decided to accompany her on the trip to Finland. Are there any Haskellers in the area who might be interested in an informal meetup? I'd also welcome any suggestions on what to see, do, eat, etc. We'll have 3-4 days after her conference. We especially enjoy nature (hiking, biking, boating), music (folk/ethnic, jazz, classical), and dance (performance and participatory). And I love Haskell, of course. As my inquiry is only weakly Haskell-related, I suggest that replies be sent to me rather than the list. Thanks. Dean (from North Carolina)

1 0

pointer equality
by Nikhil A. Patil 21 Jul '11

21 Jul '11

Hi, Is there any way of getting the following code to immediately return True without performing the element-by-element comparison? Essentially this boils down to checking whether pointers are equal before comparing the contents. > main = print $ f == f > where f = [1..10^9] Thanks!! nikhil

19 25

haskell zlip read position
by rahul 21 Jul '11

21 Jul '11

Hi Guys, I am trying to parse a binary stream with the format for one entry [headers, zlib compressed content] , with multiple entries. I can use the Zlib library to get the content for the first entry after the headers, but I cannot find a way to get the offset to start parsing for the second entry. Is there a way I can get this information out of ZLib? or is there a better approach to doing this? Any pointers would be very much appreciated. Regards, Rahul

3 6

Network.CGI and SQL Queries
by william murphy 21 Jul '11

21 Jul '11

Hello there, There's an odd error every time I try to run this code: import Network.CGI import Text.XHtml import qualified Text.Html as H import Database.HDBC import Database.HDBC.Sqlite3 import System.Random ... sqlReaderIO = do handle <- connectSqlite3 "table2.db" scoresSql <- quickQuery' handle "SELECT * FROM pachinkoScores ORDER BY score, name" [] let scores = map (map (\x -> fromSql x :: String)) scoresSql return (show scores) ... cgiTMain = do ... scoresPassed <- sqlReaderIO output . renderHtml $ (inputForm n ss) maybe (inputForm n ss) (\mT -> page "Loading..." (visual t v a mT ss) n v a (t + 1) mT) maxT main = runCGI $ handleErrors cgiTMain When I remove the "scoresPassed <- sqlReaderIO" line, the code run smoothly. Nothing else references "scoresPassed". The ghc, however, gives me this error when the line is included: tankwSql.hs:144:15: No instance for (MonadCGI IO) arising from a use of `output' Possible fix: add an instance declaration for (MonadCGI IO) In the first argument of `(.)', namely `output' In the expression: output . renderHtml In the expression: output . renderHtml $ inputForm n tankwSql.hs:147:30: Couldn't match expected type `CGIT m0 CGIResult' with actual type `IO CGIResult' In the first argument of `handleErrors', namely `cgiTMain' In the second argument of `($)', namely `handleErrors cgiTMain' In the expression: runCGI $ handleErrors cgiTMain Thanks, Will

3 2

Re: [Haskell-cafe] pointer equality
by wren ng thornton 20 Jul '11

20 Jul '11

On 7/20/11 1:22 PM, Chris Smith wrote: > If the latter, then it seems this would be a > pretty serious garbage collector bug, and that it would be impossible > that such a bug wouldn't also break other code that doesn't use pointer > equality at all. After all, we've got a running user thread, which if > it were to force those thunks now they would necessarily be observably > equal, but if it doesn't and waits until later they may be different? Not at all. Some amount of time passes between obtaining the pointer of x and obtaining the pointer of y in order to test whether they're equal. Given that time passes between these two steps, it's possible for GC to happen after getting one of the pointers and then when returning to user code we get the other pointer. It so happens that GC will never happen at that point on current versions of GHC, but that's not to say it couldn't (as it mustn't if the API specified no false positives). I can't think of any plausible optimization that would lead to a design which allows false positives, but I'm not sure pointer equality should really be offered in the first place. That's what StableNames are for. Pointer equality may be a reasonable implementation of StableName equality on some systems, but I really think that the two ideas should be kept separate. -- Live well, ~wren

1 0

Re: [Haskell-cafe] pointer equality
by wren ng thornton 20 Jul '11

20 Jul '11

On 7/20/11 5:08 PM, Paul Johnson wrote: > I would have thought that the compiler, as a matter of optimisation, > could insert a check to see if (==) is comparing an object with itself. > The only way I can see this breaking is with perverse instances of Eq > that would return False for "f == f". Like the instance for Double (where f = NaN)? -- Live well, ~wren

1 0