December 2010 - Haskell-Cafe

non_lazy_ptr ld error?
by Evan Laforge 23 Dec '10

23 Dec '10

Lately I've been seeing this error occasionally when recompiling: Undefined symbols: "_UiziTrackC_d26DB", referenced from: _UiziTrackC_d26DB$non_lazy_ptr in TrackC.o ld: symbol(s) not found If I delete the .o file for the mentioned module and recompile, it links fine. This makes me think something in the dependency analysis is mixed up and not recompiling something that should be recompiled. But I'm using --make and letting ghc figure out the deps. Does this error look familiar to anyone? This is ghc 6.12.3 on OS X 10.5.8. On a probably unrelated note, a few times lately I've had a startup segfault on a program which was previously fine. The stack trace on the crashing thread is: Thread 2 Crashed: 0 libSystem.B.dylib 0x954f9fa0 strcmp + 80 1 elaforge.seq.seq 0x0143f7dd lookupHashTable + 131 2 elaforge.seq.seq 0x01444243 ghciInsertStrHashTable + 24 3 elaforge.seq.seq 0x01445a87 ocGetNames_MachO + 1803 4 elaforge.seq.seq 0x014447bf loadObj + 500 5 elaforge.seq.seq 0x007064a4 sYy_info + 12 This is a little worrying because I've never seen this before and I don't know why it's started. It seems to happen consistently until I make clean and recompile. My app uses hint and I suspect this is part of the hint initialization and has to do with inconsistent .o files or something... anyone seen this either?

1 0

Yet Another Haskell Tutorial missing (maybe lost on server move?)
by Henning Thielemann 23 Dec '10

23 Dec '10

The page http://www.haskell.org/haskellwiki/Yet_Another_Haskell_Tutorial says that there is a darcs repository at http://darcs.haskell.org/yaht/ but it is not. Also in the canonical replacement location http://code.haskell.org/yaht/ I cannot find it. The Wikiboook https://secure.wikimedia.org/wikibooks/en/wiki/Haskell/YAHT mentions the same URL and additionally a PDF at http://darcs.haskell.org/yaht/yaht.pdf which is missing, too. Can someone please tell me, where it is, update HaskellWiki and Wikibook and add a forward from darcs.haskell.org to the new location?

1 0

New simplified primes; no VIP necessary.
by Will Ness 23 Dec '10

23 Dec '10

Hi, For those who remember the discussion about this about a year ago, it turns out there was a simpler version after all lurking somewhere in there (or is it _out_?). I've just posted it to the haskellwiki's Prime Numbers page: primes = 2: primes' where primes' = 3: 5: [7,9..] `minus` tfold [ [p*p,p*p+2*p..] | p <- primes' ] tfold ((x:xs):t) = x : xs `union` tfold (pairs t) pairs ((x:xs):ys:t) = (x: union xs ys) : pairs t The full code with double-feed, wheel and better folding structure is also there. Speed and complexity weren't changed by this. But it is rather _short_. :) Cheers,

1 0

ANN: cabal-dev 0.7.3.1 -- now with ghci!
by Rogan Creswick 23 Dec '10

23 Dec '10

Cabal-dev is now capable of launching ghci with the project's package database and local modules (if the package under development exposes a library). For example: # First, invoke cabal-dev install the package to populate the # package database: $ cabal-dev install .... <snip> .... $ cabal-dev ghci GHCi, version 6.12.3: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer-gmp ... linking ... done. Loading package base ... linking ... done. Loading package ffi-1.0 ... linking ... done. Prelude> The ghci shell should have access to all the libraries your application/library is using, as well as any modules that your library exposes. Note that this is not quite as natural as your traditional ghci shell, namely: Source modifications are not visible without exiting, re-issuing `cabal-dev install` *and* `cabal-dev ghci`. This will eventually get better, but that's where things are right now. The reason for this is that `cabal-dev ghci` just issues ghci with the cabal-dev package database (and excluding the user package db, to best reflect what cabal-dev does when it causes compilation). Cabal-dev is available on hackage and github: - http://hackage.haskell.org/package/cabal-dev - https://github.com/creswick/cabal-dev There is a short write-up with a (slightly) longer example here: - http://blog.ciscavate.org/2010/12/cabal-dev-sandboxing-your-haskell-develop… --Rogan

1 0

Behaviour of System.Directory.getModificationTime
by Arnaud Bailly 22 Dec '10

22 Dec '10

Hello, I am writing a program which scans a bunch of directories to detect changes (added, modified, deleted files). To detect modification, I would like to use getModification, doing something like the following: -- |Stores last modified timestamp of the file -- Yes, I come from Java... type HashMap = M.Map FilePath ClockTime data (Show a, Eq a) => Edit a = Mod a | Add a | Del a deriving (Eq,Show) checkChanges :: [String] -> HashMap -> IO ([Edit FilePath], HashMap) checkChanges [fs] m = do addedFilesList <- lsRecursive fs timestamps <- mapM getModificationTime addedFilesList let allts = zip addedFilesList timestamps let ret = (findDeletedFiles allts.updateScannedFiles allts) ([],m) return ret -- | Returns an updated map and a list of modified/added/deleted files updateScannedFiles :: [(FilePath,ClockTime)] -> ([Edit FilePath], HashMap) -> ([Edit FilePath], HashMap) updateScannedFiles [] r = r updateScannedFiles ((path,ts):files) (updates,m) = case M.lookup path m of Nothing -> updateScannedFiles files ((Add path:updates), M.insert path ts m) Just ts' -> if ts' < ts then updateScannedFiles files ((Mod path:updates), M.adjust (const ts) path m) else updateScannedFiles files (updates, m) -- omitting findDeletedFiles which is obvious and works I got the following test case that fails: "recursively marks changed files as modified" `for` do complexFileSetup -- create a small FS tree with 3 files root <- tempDir state <- checkChanges [root] M.empty writeFile (root </> "subdir" </> "cFile.txt") "this is another toast" checkChanges [root] (snd state) >>= (\r -> assertEqual "there should be 1 changed file in 3 files" 1 ((length.modified.fst) r)), In words: the getModificationTime of the overwritten file appears identical before and after the write. What am I doing wrong ? BTW, I checked the content of the file which has been correctly updated. Thanks for your help Arnaud

4 5

$ do?
by Jonathan Geddes 22 Dec '10

22 Dec '10

Quick question: Why do I need the $ in the following bits of code? > main = withSocketsDo $ do > --do something with sockets > foo = fromMaybe 0 $ do > --do something in the maybe monad I don't see (after admittedly only a minute or so thinking about it) where any grammar ambiguities would be if 'do' had an implicit $ in front of it: > foo = fromMaybe 0 do > --do something in maybe Though now that I've written it down, that is hard for me to visually parse at a glance. I'm still curious though. --Jonathan

5 4

Re: [Haskell-cafe] Proof in Haskell
by Patrick Browne 22 Dec '10

22 Dec '10

On 22/12/2010 14:48, Artyom Shalkhakov wrote: > ..Do you want to prove a property of > a function formally, using some kind of formal logic? I am aware that functional languages do not do proofs at term level, but the motivation for my question is to get a precise reason why this is so. The replies from the café have clearly articulated the reasons. Thanks to all, Pat This message has been scanned for content and viruses by the DIT Information Services E-Mail Scanning Service, and is believed to be clean. http://www.dit.ie

1 0

Darcs failure
by Andrew Coppin 22 Dec '10

22 Dec '10

I'm running a VM with Ubuntu 10.10 (Maverick Meerkat). I installed Darcs 2.4.4 using apt-get, but it keeps CONSTANTLY failing with the error message darcs: bug at src/URL.hs:246 compiled Sep 12 2010 20:24:56 Another possible bug in URL.waitNextUrl: curl_multi_perform() - no running handles See http://wiki.darcs.net/BugTracker/Reporting for help on bug reporting. Does anybody have any clue how I can get the hell rid of this stupid message and actually get some stuff done? (At this point, Darcs has hung, and I have to use the kill command with just the right signal number to make Darcs exit. Ctrl+C just tells me "Couldn't handle interrupt since darcs was in a sensitive job." So every single time this error happens, I have to get a process list, find Darcs, issue the kill command, and then reissue the original Darcs command and hope it actually completes this time...)

4 5

Identity type
by Pierre-Etienne Meunier 22 Dec '10

22 Dec '10

Hi, Is there something like an identity type, transparent to the type-checker, in haskell ? For instance, I'm defining an interval arithmetic, with polynomials, matrices, and all that... defined with intervals. The types are : Polynomial Interval (instead of Polynomial Double for instance) Matrix Interval etc. Now I'd like to be able to define some typeclass "intervalize" that would convert a IEEE-754 version of my types into the interval ones : class Intervalize a where intervalize::a Double-> a Interval Wouldn't it be very natural wanting to use this class on Doubles directly : intervalize' :: Double->Interval intervalize'=intervalize Does this exist in haskell ? in GHC ? Cheers, Pierre

2 1

Missing Parsec library in latest stable GHC
by Aaron Gray 22 Dec '10

22 Dec '10

Missing Parsec library :- scheme.o(.text+0x4fa):fake: undefined reference to `parseczm2zi1zi0zi0_TextziParserCombinatorsziParsecziCombinator_skipMany1_closure' scheme.o(.text+0x501):fake: undefined reference to `parseczm2zi1zi0zi0_TextziPaserCombinatorsziParsecziChar_space_closure' scheme.o(.text+0x5c2):fake: undefined reference to `parseczm2zi1zi0zi0_TextziParserCombinatorsziParsecziChar_oneOf_closure' scheme.o(.text+0x63a):fake: undefined reference to `mtlzm1zi1zi0zi0_ControlziMonadziTrans_zdf1_closure' GHC from :- http://www.haskell.org/ghc/dist/stable/dist/ ghc-7.0.1.20101221-i386-windows.exe<http://www.haskell.org/ghc/dist/stable/dist/ghc-7.0.1.20101221-i386-windows…> and earlier. Aaron

3 4