- Haskell-Cafe - Haskell.org

cabal Setup.hs configure - cabal doesnt produce a setup-config
by geoffrey 19 May '08

19 May '08

Hi, I am trying to install nanocurses, but cabal isn't creating the setup-config file in dist/ I have run (and re-run) runghc Setup.hs configure The first time it complained about mpg123 (whatever that is) so i installed the other dependencies - unix & bytestring Then i tried to re-run setup configure for nano again. Configure creates the dist directory but no files. I have deleted and untarred the files and retried,but to no avail. Any clues? Thanks. (OS: Ubuntu hardy, GHC: 6.8.2)

2 2

Control.Exception.evaluate - 'correct definition' not so correct
by Bryan Donlan 19 May '08

19 May '08

Hi all, After some discussion on #haskell I decided to bring this issue to haskell-cafe. GHC's documentation for Control.Exception.evaluate states: evaluate :: a -> IO a Forces its argument to be evaluated when the resultant IO action is executed. It can be used to order evaluation with respect to other IO operations; its semantics are given by evaluate x `seq` y ==> y evaluate x `catch` f ==> (return $! x) `catch` f evaluate x >>= f ==> (return $! x) >>= f Note: the first equation implies that (evaluate x) is not the same as (return $! x). A correct definition is evaluate x = (return $! x) >>= return However, if >>= is strict on its first argument, then this definition is no better than (return $! x). One might next consider: evaluate x = (return x) >>= (return $!) However, according to the monad laws, this is equivalent to: evaluate x = return $! x and we're back to where we started. Although GHC's implementation of IO is somewhat more relaxed about this, there is no guarentee that this will be the case in all IO implementations, or future versions of GHC, or different optimization flags, or... The best I've come up with so far is: evaluate x = newIORef (return $! x) >>= join . readIORef In any case, if >>= is to be guarenteed lazy, this ought to be documented somewhere (or is it?). Otherwise Control.Exception's docs should be updated to provide a more correct example and/or the caveat that >>= must not be left-strict for the example implementation to be correct. Thanks, Bryan Donlan

10 13

Two-iteration optimisation (was: GHC Predictability)
by Paul Johnson 19 May '08

19 May '08

We've had a big debate over > mean xs = foldl' (+) 0 xs / fromIntegral (length xs) For anyone who didn't follow it, the problem is that "mean" needs to traverse its argument twice, so the entire list has to be held in memory. So if "xs = [1..1000000000]" then "mean xs" uses all your memory, but "foldl' (+) 0 xs" and "length xs" by themselves do not. The solution is for the programmer to rewrite "mean" to accumulate a pair containing the running total and count together, then do the division. This makes me wonder: could there be a compiler optimisation rule for this, collapsing two iterations over a list into one. I've never tried writing GHC rules, but something like: > f (foldl' g x) (foldl' h x) = (uncurry f) (foldl' (\i (gt,ht) -> (g i gt, h i ht))) The first problem that occurs to me is the number of permutations of fold and map functions. Paul.

8 10

Question on declaring an instance
by john lask 19 May '08

19 May '08

Consider the data type: Foo i o a = ... we want to declare an instance of Monad and Arrow for Foo without using a newtype (repackaging). The effect we want is to be able to construct an expression like: ( f>> g )>>> ( h>> i ) where f, g, h, i :: Foo i o a, for some i o a it is easy to declare instance Monad (Foo i o) where ... but how do we declare instance Arrow (Foo ... ), this instance takes a kind * -> * -> * it is easy to declare instance Arrow (Bar a) where ... where Bar a i o is a flipping of the order of the arguments to Foo i.e. Bar = ((flip . (flip .)) Foo Can Foo be declared an instance of both Arrow and Monad ? _________________________________________________________________ It's simple! Sell your car for just $30 at CarPoint.com.au http://a.ninemsn.com.au/b.aspx?URL=http%3A%2F%2Fsecure%2Dau%2Eimrworldwide%…

2 1

Parsec3 performance issues (in relation to v2)
by Neal Alexander 18 May '08

18 May '08

Just a heads up - i only have a month or so experience with Haskell, so alot of these issues may be my own fault. Anyway, the log file that i'm parsing uses English grammar, and the performance really dropped just by upgrading to Parsec3. I was hoping to use the ByteString support to boost the speed of already slow code, but had no such luck. It basicly went from "Ugh, this is kinda slow" to "Uhhhh i'm gonna go grab a burger and let this melt my CPU" haha. If anything, its probably all the look-ahead the rules have to do to get the context specific stuff right. Some of the code is here: http://hpaste.org/7578 ------------------------------------------------------------- #1 . Parsec 2: total time = 46.44 secs (2322 ticks @ 20 ms) total alloc = 16,376,179,008 bytes (excl. profiling overheads) Parse taking 51.3% time and 65.3% alloc. ------------------------------------------------------------- ------------------------------------------------------------- #2 . Parsec3 (4 times slower, no code changes): total time = 181.08 secs (9054 ticks @ 20 ms) total alloc = 46,002,859,656 bytes (excl. profiling overheads) Text.Parsec.Prim Taking 84.7% time and 86.0% alloc. ------------------------------------------------------------- ------------------------------------------------------------- #3 . Parsec3 but with the whole project converted to ByteString: (8 times slower): total time = 378.22 secs (18911 ticks @ 20 ms) total alloc = 100,051,417,672 bytes (excl. overheads) ------------------------------------------------------------- The third parse probably isn't a great indicator, since i reverted some rule-set optimizations that were causing errors. Plus i ended up packing the parsec String results to ByteStrings to fit in with everything else. I can post the full profiling info if anyone really cares.

6 11

Re: [Haskell-cafe] Re: ghc on Ubuntu Linux?
by Galchin, Vasili 18 May '08

18 May '08

Hi Josef, What generates dist/setup-config? When I run "runhaskell Setup.hs configure", nothing including dist/setup.config gets generated. ?? Regards, Vasili On Sat, May 17, 2008 at 11:02 AM, Josef Svenningsson < josef.svenningsson(a)gmail.com> wrote: > On Sat, May 17, 2008 at 1:00 PM, Galchin, Vasili <vigalchin(a)gmail.com> > wrote: > > Josef, > > > > E.g. > > vigalchin@ubuntu:~/FTP/Haskell/unix-2.2.0.0$ runhaskell Setup.hs > configure > > Configuring unix-2.3.0.0... > > > > Normally copious output ... ??? > > > That's what it should look like! It just means you have a recent > version of cabal which doesn't spit out tons of information when > configuring. Happily all is well. > > /Josef >

3 2

Richer (than ascii) notation for haskell source?
by Patrick Surry 18 May '08

18 May '08

Probably a silly question, but for me one of the nice things about Haskell is that it's a lot like just writing math(s). But in contrast to math you lose a lot of notational flexibility being limited to the ascii character set in your source code. It would be nice to be able to use a richer set of symbols in your source code for operators and functions (e.g. integral, sum, dot and cross-product, ...), as well as variables (the standard upper and lower-case greek for example, along with things like super- and sub-scripting, bold/italic and what-not). You could imagine ending up with source code that reads just like a mathematical paper. Don't know how I'd actually then write/maintain the source-code - some WYSIWYG editor or effectively writing it in '(la)tex'? Maybe that's what Knuth is on about with his 'literate programming' weave/tangle stuff which I don't know much about - does that translate to Haskell? Patrick Patrick.Surry(a)portraitsoftware.com <mailto:Patrick.Surry@portraitsoftware.com> , VP Technology Tel: (617) 457 5230 Mob: (857) 919 1700 Fax: (617) 457 5299 Map <http://maps.google.com/maps?f=q&hl=en&q=125+Summer+St,+Boston,+MA+02110 &ie=UTF8&z=15&ll=42.353153,-71.057296&spn=0.022644,0.054245&om=1&iwloc=A > Portrait Software introduces Portrait Campaign Manager <http://www.portraitsoftware.com/Products/portrait_campaign_manager> : Easy, integrated campaign management for automated and highly targeted outbound marketing http://supportcentre.portraitsoftware.com/Vmail/emailfooter/balloon.gif <http://www.portraitsoftware.com/> http://supportcentre.portraitsoftware.com/Vmail/emailfooter/portrait_sof tware_logo.gif <http://www.portraitsoftware.com/> www.portraitsoftware.com <http://www.portraitsoftware.com/> ________________________________________________________________________ DISCLAIMER: This e-mail is intended only for the addressee named above. As this e-mail may contain confidential or privileged information, if you are not the named addressee, you are not authorised to retain, read, copy or disseminate this message or any part of it. If you received this email in error, please notify the sender and delete the message from your computer.

17 39

Write Haskell as fast as C. [Was: Re: GHC predictability]
by Don Stewart 18 May '08

18 May '08

> I offer up the following example: > > mean xs = sum xs / length xs > > Now try, say, "mean [1.. 1e9]", and watch GHC eat several GB of RAM. (!!) > > If we now rearrange this to > > mean = (\(s,n) -> s / n) . foldr (\x (s,n) -> let s' = s+x; n' = n+1 > in s' `seq` n' `seq` (s', n')) (0,0) > > and run the same example, and watch it run in constant space. > > Of course, the first version is clearly readable, while the second one > is almost utterly incomprehensible, especially to a beginner. (It's even > more fun that you need all those seq calls in there to make it work > properly.) > > The sad fact is that if you just write something in Haskell in a nice, > declarative style, then roughly 20% of the time you get good > performance, and 80% of the time you get laughably poor performance. I've written an extended post on how to understand and reliably optimise code like this, looking at it all the way down to the assembly. The result are some simple rules to follow for generated code as good as gcc -O2. Enjoy, http://cgi.cse.unsw.edu.au/~dons/blog/2008/05/16#fast -- Don

16 32

ByteString comparison question (was: another Newbie performance question)
by Philip Müller 18 May '08

18 May '08

Hi, I'm changing my CSV program to use ByteStrings, but I have problems with this: readCSVLine :: String -- line as String -> [String] -- line broken down into the value Strings readCSVLine = unfoldr builder where builder [] = Nothing builder xs = Just $ readField xs readField [] = ([],[]) readField (',':xs) = ([],xs) readField ('"':xs) = (field,rest) where (field,'"':rest) = break (== '"') xs So far, I have something like this - doesn't look too good to me, and doesn't compile: import qualified Data.ByteString as B import qualified Data.ByteString.Char8 as C8 type CSV = [[B.ByteString]] readCSVLine :: B.ByteString -- line as String -> [B.ByteString] -- line broken down into the value Strings readCSVLine = unfoldr builder where builder xs | xs == B.empty = Nothing | otherwise = Just $ readField xs readField xs | xs == B.empty = ([],[]) | B.head xs == ',' = ([], B.tail xs) | B.head xs == '"' = (field, B.tail rest) where field,rest) = B.break (== '"') xs I do not know how to compare a Word8 to a Char. Or maybe I don't need to? Regards Philip

3 4

Type inference; always
by Peter Verswyvelen 18 May '08

18 May '08

In Haskell, I sometimes have to annotate code with type info because the type-inferer otherwise fails (even with |-XNoMonomorphismRestriction)|. Surely, most of the time this is because I was writing buggy code, but sometimes, type annotation just seems needed to get a successful compilation (luckily not as often as in C# 3.0) It seems some new papers appeared regarding this, e.g. http://research.microsoft.com/~simonpj/papers/boxy Is this work being incorporated into GHC? Unfortunately I'm not able to read those papers, so I'm not really sure what it means anyway and what implications (example code?) it would have ;-) Cheers, Peter

5 5